Assessment report - European Medicines Agency · MoA Mechanism of Action MPA Microscopic...
Transcript of Assessment report - European Medicines Agency · MoA Mechanism of Action MPA Microscopic...
30 Churchill Place ● Canary Wharf ● London E14 5EU ● United Kingdom
An agency of the European Union
Telephone +44 (0)20 3660 6000 Facsimile +44 (0)20 3660 5555
Send a question via our website www.ema.europa.eu/contact
© European Medicines Agency, 2017. Reproduction is authorised provided the source is acknowledged.
21 April 2017 EMA/303207/2017 Committee for Medicinal Products for Human Use (CHMP)
Assessment report
Rixathon
International non-proprietary name: rituximab
Procedure No. EMEA/H/C/003903/0000
Note
Assessment report as adopted by the CHMP with all information of a commercially confidential nature
deleted.
Assessment report
EMA/303207/2017 Page 2/115
Table of contents
1. Background information on the procedure .............................................. 7
1.1. Submission of the dossier ...................................................................................... 7
1.2. Steps taken for the assessment of the product ......................................................... 9
2. Scientific discussion .............................................................................. 10
2.1. Problem statement ............................................................................................. 10
2.1.1. Disease or condition ......................................................................................... 10
2.1.2. Epidemiology .................................................................................................. 11
2.1.3. Biologic features .............................................................................................. 11
2.1.4. Clinical presentation and diagnosis .................................................................... 12
2.2. Quality aspects .................................................................................................. 14
2.2.1. Introduction .................................................................................................... 14
2.2.2. Active Substance ............................................................................................. 14
2.2.3. Finished Medicinal Product ................................................................................ 17
2.2.4. Discussion on chemical, pharmaceutical and biological aspects .............................. 22
2.2.5. Conclusions on the chemical, pharmaceutical and biological aspects ...................... 23
2.2.6. Recommendation(s) for future quality development ............................................. 23
2.3. Non-clinical aspects ............................................................................................ 23
2.3.1. Introduction .................................................................................................... 23
2.3.2. Pharmacology ................................................................................................. 23
2.3.3. Pharmacokinetics............................................................................................. 27
2.3.4. Toxicology ...................................................................................................... 28
2.3.5. Ecotoxicity/environmental risk assessment ......................................................... 30
2.3.6. Discussion on non-clinical aspects...................................................................... 30
2.3.7. Conclusion on the non-clinical aspects ................................................................ 32
2.4. Clinical aspects .................................................................................................. 32
2.4.1. Introduction .................................................................................................... 32
2.4.2. Pharmacokinetics............................................................................................. 33
2.4.3. Pharmacodynamics .......................................................................................... 47
2.4.4. Discussion on clinical pharmacology ................................................................... 52
2.4.5. Conclusions on clinical pharmacology ................................................................. 55
2.5. Clinical efficacy .................................................................................................. 55
2.5.1. Dose response studies...................................................................................... 55
2.5.2. Main study ...................................................................................................... 55
2.5.3. Discussion on clinical efficacy ............................................................................ 79
2.5.4. Conclusions on the clinical efficacy ..................................................................... 83
2.6. Clinical safety .................................................................................................... 84
2.6.1. Discussion on clinical safety .............................................................................. 97
2.6.2. Conclusions on the clinical safety ..................................................................... 100
2.7. Risk Management Plan ...................................................................................... 101
2.8. Pharmacovigilance ............................................................................................ 107
2.9. Product information .......................................................................................... 107
2.9.1. User consultation ........................................................................................... 107
Assessment report
EMA/303207/2017 Page 3/115
2.9.2. Additional monitoring ..................................................................................... 107
3. Benefit-Risk Balance............................................................................ 108
3.1. Therapeutic Context ......................................................................................... 108
3.1.1. Disease or condition ....................................................................................... 108
3.1.2. Available therapies and unmet medical need ..................................................... 108
3.1.3. Main clinical studies ....................................................................................... 108
3.2. Favourable effects ............................................................................................ 108
3.3. Uncertainties and limitations about favourable effects ........................................... 109
3.4. Unfavourable effects ......................................................................................... 109
3.5. Uncertainties and limitations about unfavourable effects ....................................... 110
3.6. Effects Table .................................................................................................... 111
3.7. Benefit-risk assessment and discussion ............................................................... 111
3.7.1. Importance of favourable and unfavourable effects ............................................ 111
3.7.2. Balance of benefits and risks ........................................................................... 111
3.7.3. Additional considerations on the benefit-risk balance ......................................... 111
3.8. Conclusions ..................................................................................................... 112
4. Recommendations ............................................................................... 112
Assessment report
EMA/303207/2017 Page 4/115
List of abbreviations
AC Acceptance Criteria
ACR (20 /50/ 70) American College of Rheumatology 20% (50%) (70%) Response Criteria
ADA Anti-drug antibody
ADCC Antibody dependent cellular cytotoxicity
ADCP Antibody-Dependent cellular phagocytosis
ADR Adverse drug reaction
AE Adverse event
AF4 Asymmetrical flow field-flow fractionation
Asn Asparagine
AUC Area under the concentration-time curve
AUC Analytical ultracentrifugation
AUC(0-inf) Area under the serum concentration-time curve from time zero to infinity
AUC(0-last) AUC calculated from start of dose to the end of the dosing interval, tau
AUCall AUC from the time of dosing to the time of the last observation
AUClast AUC from time zero to the last measured time point
AUEC Area under the effect-time curve
AUEC(0-t) The area under the effect-time curve from time zero to time ‘t’.
BSA Body surface area
CDAI Clinical Disease Activity Index
CDC Complement-dependent cytotoxicity
CEX Cation Exchange Chromatography
CHO Chinese Hamster Ovary
CHOP Cyclophosphamide, hydroxydaunorubicin, oncovin [vincristine] and prednisone
CI Confidence interval(s)
CLL Chronic lymphocytic leukemia
Cmax The maximum (peak) observed serum concentration of rituximab
Cmin Minimum observed concentration
CPP Critical process parameter
CR Complete response
CRP C-reactive protein
CT Computerized tomography
CTCAE Common Terminology Criteria for Adverse Events
Ctrough concentration before the next infusion dose administration
CVP Cyclophosphamide, vincristine, prednisone
Da Dalton
DAS Disease Activity Score
DAS28 Disease Activity Score Based On A 28 Joint Count
DLBCL Diffuse large B-cell lymphoma
DMARD Disease modifying anti-rheumatic drug
DNA Deoxyribonucleic acid
DSC Differential Scanning Calorimetry
DSP Down Stream Process
ECG Electrocardiogram
ECL Electrochemiluminescence
ECOG Eastern Cooperative Oncology Group
Assessment report
EMA/303207/2017 Page 5/115
ELISA Enzyme-linked immunosorbent assay
ESR Erythrocyte sedimentation rate
EULAR European League Against Rheumatism
FAS Full Analysis Set
Fc Fragment crystallisable (region of an antibody)
FL Follicular lymphoma
FLIPI Follicular Lymphoma International Prognostic Index
FMEA Failure Mode Effects Analysis
FP Finished product
FT-IR Fourrier Transform Infrared Spectroscopy
GCP Good clinical practice
GMP Good Manufacturing Practice
GPA Granulomatosis with polyangiitis
HAQ-DI Health Assessment Questionnaire – Disability Index
HBV Hepatitis B virus
HCP Health Care Provider
HDX Hydrogen Deuterium Exchange
HIV Human immunodeficiency virus
HPLC High performance liquid chromatography
HR Hazard ratio
i.v. Intravenous
ICH International Conference on Harmonisation of Technical Requirements for
Registration of Pharmaceuticals for Human Use
Ig Immunoglobulin
INN International Nonproprietary Name
IPC In-process Control
IRR Infusion-related reaction
ITT Intention-to-Treat
LLOQ Lower limit of quantification
mAb Monoclonal antibody
MALLS Multi-angle Laser Light Scattering
MCB Master Cell Bank
MedDRA Medical Dictionary for Regulatory Activities
MFI Micro-flow imaging
mg Milligram
mL Millilitre
MoA Mechanism of Action
MPA Microscopic polyangiitis
MRI Magnetic Resonance Imaging
MTX Methotrexate
NAb Neutralizing antibody
NHL Non-Hodgkin’s Lymphoma
NMR Nuclear Magnetic Resonance
ORR Overall response rate
OS Overall survival
PC Process characterization
PD Pharmacodynamics
PD Progressive disease
Assessment report
EMA/303207/2017 Page 6/115
PFS Progression free survival
Ph. Eur. European Pharmacopoeia
PI Package Insert
PK Pharmacokinetics
PML Progressive multifocal leukoencephalopathy
PP Process parameter
PPS Per Protocol Set
PR Partial response
PT Preferred term
PTM Post-translational modifications
QbD Quality by design
RA Rheumatoid arthritis
RF Rheumatoid factor
SA Scientific Advice
SAE Serious adverse event
SD Standard deviation
SD Stable disease
SDAI Simplified Disease Activity Index
SEC Size-exclusion chromatography
SOC System organ class
SPR Surface plasmon resonance
T1/2 Elimination half-life
TEAE Treatment emergent adverse event
Tmax The time to reach maximum (peak) serum concentration after single dose
TNF Tumour necrosis factor
TSE Transmissible Spongiform Encephalopathy
UF/DF Ultrafiltration/diafiltration
US The United States of America
UV Ultraviolet
WCB Working Cell Bank
Assessment report
EMA/303207/2017 Page 7/115
1. Background information on the procedure
1.1. Submission of the dossier
The applicant Sandoz GmbH submitted on 11 April 2016 an application for marketing authorisation to
the European Medicines Agency (EMA) for Rixathon, through the centralised procedure falling within
the Article 3(1) and point 1 of Annex of Regulation (EC) No 726/2004. The eligibility to the centralised
procedure was agreed upon by the EMA/CHMP on 21 November 2013.
The applicant applied for the following indication:
Rixathon is indicated in adults for the following indications:
Non-Hodgkin’s lymphoma (NHL)
Rixathon is indicated for the treatment of previously untreated patients with stage III-IV follicular
lymphoma in combination with chemotherapy.
Rixathon maintenance therapy is indicated for the treatment of follicular lymphoma patients
responding to induction therapy.
Rixathon monotherapy is indicated for treatment of patients with stage III-IV follicular lymphoma who
are chemoresistant or are in their second or subsequent relapse after chemotherapy.
Rixathon is indicated for the treatment of patients with CD20 positive diffuse large B cell non-
Hodgkin’s lymphoma in combination with CHOP (cyclophosphamide, doxorubicin, vincristine,
prednisolone) chemotherapy.
Chronic lymphocytic leukaemia (CLL)
Rixathon in combination with chemotherapy is indicated for the treatment of patients with previously
untreated and relapsed/refractory chronic lymphocytic leukaemia. Only limited data are available on
efficacy and safety for patients previously treated with monoclonal antibodies including rituximab or
patients refractory to previous rituximab plus chemotherapy.
Rheumatoid arthritis
Rixathon in combination with methotrexate is indicated for the treatment of adult patients with severe
active rheumatoid arthritis who have had an inadequate response or intolerance to other disease
modifying anti-rheumatic drugs (DMARD) including one or more tumour necrosis factor (TNF) inhibitor
therapies.
Rituximab has been shown to reduce the rate of progression of joint damage as measured by X-ray
and to improve physical function, when given in combination with methotrexate.
Granulomatosis with polyangiitis and microscopic polyangiitis
Rixathon, in combination with glucocorticoids, is indicated for the induction of remission in adult
patients with severe, active granulomatosis with polyangiitis (Wegener’s) (GPA) and microscopic
polyangiitis (MPA).
Assessment report
EMA/303207/2017 Page 8/115
The legal basis for this application refers to:
Article 10(4) of Directive 2001/83/EC – relating to applications for a biosimilar medicinal products. The
application submitted is composed of administrative information, complete quality data, appropriate
non-clinical and clinical data for a similar biological medicinal product.
The chosen reference product is:
Medicinal product which is or has been authorised in accordance with Community provisions in force for
not less than 6/10 years in the EEA:
Product name, strength, pharmaceutical form: MabThera, 100mg and 500mg, concentrate for solution for infusion
Marketing authorisation holder: Roche Registration Ltd
Date of authorisation: 02-06-1998
Marketing authorisation granted by:
Community
Community Marketing authorisation number: EU/1/98/067/001-002
Medicinal product authorised in the Community/Members State where the application is made or
European reference medicinal product:
Product name, strength, pharmaceutical form: MabThera, 100mg and 500mg, concentrate for solution for infusion
Marketing authorisation holder: Roche Registration Ltd
Date of authorisation: 02-06-1998
Marketing authorisation granted by:
Community
Community Marketing authorisation number: EU/1/98/067/001-002
Medicinal product which is or has been authorised in accordance with Community provisions in force
and to which bioequivalence has been demonstrated by appropriate bioavailability studies:
Product name, strength, pharmaceutical form: MabThera, 100mg and 500mg, concentrate for solution for infusion
Marketing authorisation holder: Roche Registration Ltd
Date of authorisation: 02-06-1998
Marketing authorisation granted by:
Community
Community Marketing authorisation number(s): EU/1/98/067/001-002
Information on Paediatric requirements
Not applicable.
Assessment report
EMA/303207/2017 Page 9/115
Information relating to orphan market exclusivity
Similarity
Pursuant to Article 8 of Regulation (EC) No. 141/2000 and Article 3 of Commission Regulation (EC) No
847/2000, the applicant did submit a critical report addressing the possible similarity with authorised
orphan medicinal products.
Scientific Advice
The applicant received Scientific Advice from the CHMP on 25 June 2009 and 20 January 2011. The
Scientific Advice pertained to quality, non-clinical and clinical aspects of the dossier.
1.2. Steps taken for the assessment of the product
The Rapporteur and Co-Rapporteur appointed by the CHMP were:
Rapporteur: Jan Mueller-Berghaus Co-Rapporteur: Paula Boudewina van Hennik
The application was received by the EMA on 11 April 2016.
The procedure started on 19 May 2016.
The Rapporteur's first Assessment Report was circulated to all CHMP members on 4 August 2016.
The Co-Rapporteur's first Assessment Report was circulated to all CHMP members on 5 August
2016. The PRAC Rapporteur's first Assessment Report was circulated to all PRAC members on 19
August 2016.
During the meeting on 15 September 2016, the CHMP agreed on the consolidated List of Questions
to be sent to the applicant. The final consolidated List of Questions was sent to the applicant on 16
September 2016.
The applicant submitted the responses to the CHMP consolidated List of Questions on 22 December
2016.
The Rapporteurs circulated the Joint Assessment Report on the applicant’s responses to the List of
Questions to all CHMP members on 30 January 2017.
During the PRAC meeting on 9 February 2017, the PRAC agreed on the PRAC Assessment Overview
and Advice to CHMP. The PRAC Assessment Overview and Advice was sent to the applicant on 9
February 2017.
During the CHMP meeting on 23 February 2017, the CHMP agreed on a list of outstanding issues to
be addressed in writing by the applicant.
The CHMP adopted a report on similarity for Rixathon with Mabthera, Gazyvaro, Arzerra and
Imbruvica on 23 February 2017.
The applicant submitted the responses to the CHMP consolidated List of Outstanding Issues on 21
March 2017.
The Rapporteurs circulated the Joint Assessment Report on the applicant’s responses to the List of
Outstanding Issues to all CHMP members on 7 April 2017.
Assessment report
EMA/303207/2017 Page 10/115
During the meeting on 18-21 April 2017, the CHMP, in the light of the overall data submitted and
the scientific discussion within the Committee, issued a positive opinion for granting a marketing
authorisation to Rixathon on 21 April 2017.
2. Scientific discussion
2.1. Problem statement
This application concerns a centralised procedure for marketing authorisation of Rixathon, rituximab
concentrate for solution for intravenous infusion of 100 mg and 500 mg, as a biosimilar product to the
European reference product MabThera (EU/1/98/067/001-002).
MabThera has been registered for the treatment of non-Hodgkin lymphoma (NHL), chronic lymphatic
leukaemia (CLL), rheumatoid arthritis (RA), and granulomatosis with polyangiitis (GPA) and
microscopic polyangiitis (MPA). Mabthera was first authorised in the European Union in 1998.
Marketing authorisation has not been applied for solution(s) for subcutaneous injection for Rixathon
compared to the European reference product.
2.1.1. Disease or condition
Rituximab was first authorised in the European Union on 2 June 1998 under the name of MabThera. It
is also marketed under the name Rituxan in the United States (US). It is currently approved for the
following indications:
Non-Hodgkin’s lymphoma (NHL)
treatment of previously untreated patients with stage III-IV follicular lymphoma in combination
with chemotherapy.
maintenance therapy for the treatment of follicular lymphoma patients responding to induction
therapy.
monotherapy for treatment of patients with stage III-IV follicular lymphoma who are chemo-
resistant or are in their second or subsequent relapse after chemotherapy.
treatment of patients with CD20 positive diffuse large B cell non-Hodgkin’s lymphoma in
combination with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisolone) chemotherapy.
Chronic lymphocytic leukaemia (CLL)
in combination with chemotherapy for the treatment of patients with previously untreated and
relapsed/refractory chronic lymphocytic leukaemia.
Rheumatoid arthritis
in combination with methotrexate is indicated for the treatment of adult patients with severe active
rheumatoid arthritis who have had an inadequate response or intolerance to other disease-
modifying anti-rheumatic drugs (DMARD) including one or more tumour necrosis factor (TNF)
inhibitor therapies.
Granulomatosis with polyangiitis and microscopic polyangiitis
Assessment report
EMA/303207/2017 Page 11/115
in combination with glucocorticoids, is indicated for the induction of remission in adult patients with
severe, active granulomatosis with polyangiitis (Wegener’s) (GPA) and microscopic polyangiitis
(MPA).
The conditions covered by the above indications have been extensively analysed through the
respective approval procedures of Mabthera (see Mabthera European assessment report – EPAR)
2.1.2. Epidemiology
According to the prevalence of the indications Non-Hodgkin´s lymphoma (NHL), chronic lymphocytic
leukaemia (CLL) and Granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA) are
rare conditions.
The more common condition, rheumatoid arthritis, has a prevalence of more than 1 in 1000.
2.1.3. Biologic features
Non-Hodgkin’s lymphoma (NHL) is a form of malignant lymphoma distinguished from Hodgkin's
disease only by the absence of binucleate giant cells. Follicular lymphomas are indolent (slow-growing)
NHL and the second-most-common form of non-Hodgkin's lymphomas overall, defined as a lymphoma
of follicle center B-cells (centrocytes and centroblasts), which has at least a partially follicular pattern.
It is positive for the B-cell markers CD10, CD19, CD22, and usually CD20, but almost always negative
for CD5.
B-cell chronic lymphocytic leukaemia (B-CLL), also known as chronic lymphoid leukaemia (CLL), is the
most common type of leukaemia (a type of cancer of the white blood cells) in adults. CLL affects B
cell lymphocytes, which originate in the bone marrow, develop in the lymph nodes, and normally fight
infection by producing antibodies. In CLL, B cells grow in an uncontrolled manner and accumulate in
the bone marrow and blood, where they crowd out healthy blood cells.
B-cells also play several important roles in the pathogenesis of rheumatoid arthritis (RA),
granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA). They produce auto-
antibodies such as Rheumatoid Factor (RF), anti-cyclic citrullinated protein (anti-CCP) antibody in RA
or anti-neutrophil cytoplasmic antibody (ANCA) in MPA and CPA. In the synovium, RF immune
complexes may mediate complement activation and the propagation of the inflammatory cascade. B-
cells present in the RA synovial membrane may secrete a range of pro-inflammatory cytokines, some
of which are components in the process leading to joint inflammation and damage, or to induce
leukocyte infiltration. B-cells can function as antigen-presenting cells and immune-regulatory cells,
leading to T-cell activation. They can also stimulate osteoclasts and synovial fibroblasts and lead to
bone erosions and joint tissue remodelling.
Assessment report
EMA/303207/2017 Page 12/115
Peripheral B cell counts decline below normal following completion of the first dose of rituximab. In
patients treated for haematological malignancies, B cell recovery began within 6 months of
treatment and generally returned to normal levels within 12 months after completion of therapy,
although in some patients this may take longer (up to a median recovery time of 23 months post-
induction therapy). In rheumatoid arthritis patients, immediate depletion of B cells in the peripheral
blood was observed following two infusions of 1000 mg rituximab separated by a 14 day interval.
Peripheral blood B cell counts begin to increase from week 24 and evidence for repopulation is
observed in the majority of patients by week 40, whether rituximab was administered as
monotherapy or in combination with methotrexate. A small proportion of patients had prolonged
peripheral B cell depletion lasting 2 years or more after their last dose of rituximab. In patients with
granulomatosis with polyangiitis or microscopic polyangiitis, the number of peripheral blood B cells
decreased to <10 cells/μL after two weekly infusions of rituximab 375 mg/m2, and remained at that
level in most patients up to the 6 month time point. The majority of patients (81%) showed signs of
B cell return, with counts >10 cells/μL by month 12, increasing to 87% of patients by month 18.
2.1.4. Clinical presentation and diagnosis
Clinical presentation of the conditions covered by rituximab as well as diagnostic methods available
have been extensively described in the individual applications throughout the history of the originator
rituximab (Mabthera) in the EU (see Mabthera EPAR).
About the product
Rituximab is a chimeric human-murine immunoglobulin G1 (IgG1) monoclonal antibody that binds
specifically to the transmembrane antigen, CD20, a non-glycosylated phosphoprotein, located on pre-B
and mature B lymphocytes. CD20 is located on pre-B and mature B-cells, but not on haematopoietic
stem cells, pro-B-cells, normal plasma cells or other normal cells. CD20 is also expressed on >95% of
all B-cells in non-Hodgkin lymphoma and in nearly all cases in CLL. This antigen does not internalise
upon antibody binding and is not shed from the cell surface. CD20 does not circulate in the plasma as
a free antigen and, thus, does not compete for antibody binding. CD20 regulates an early step in the
activation process for cell cycle initiation and differentiation, and possibly functions as a calcium ion
channel. After binding to the CD20 antigen on the cell surface, rituximab exerts its therapeutic effect
by promoting B-cell lysis.
Type of Application and aspects on development
This application concerns a centralised procedure for marketing authorisation of Rixathon (also referred
to as GP2013), rituximab concentrate for solution for intravenous infusion of 100 mg and 500 mg, as a
biosimilar product to the European reference product MabThera (EMA registration numbers
EU/1/98/067/001-002).
MabThera has been registered for the treatment of non-Hodgkin lymphoma (NHL), chronic lymphatic
leukaemia (CLL), rheumatoid arthritis (RA), and granulomatosis with polyangiitis (GPA) and
microscopic polyangiitis (MPA). Mabthera was first authorised in the European Union in 1998.
Marketing authorisation has not been requested for solution(s) for subcutaneous injection for Rixathon
compared to the European reference product.
Assessment report
EMA/303207/2017 Page 13/115
The global development program for Rixathon was designed in a stepwise approach for demonstrating
comparability between Rixathon and the EU Reference Product MabThera. As a first step, structural
and functional characterisation of Rixathon and the reference product was established. Rixathon was
developed using the principles of a quality by design approach. A comprehensive set of binding and
activity assays were conducted to gain understanding of the functionalities, and the structures
underlying these functionalities of the molecule, that contribute to its modes of action.
The second step involved non-clinical testing including pharmacokinetics/ pharmacodynamics (PK/PD)
and toxicokinetics studies. In addition, studies in mouse xenograft tumour disease models were
conducted using a dose scaling design. Finally, assessment of effector mechanisms such as ADCC,
CDC, and apoptosis was performed in a healthy volunteer whole blood assay, and in ADCC potency
assays with different in vitro settings.
Confirmation of comparability between Rixathon and MabThera at the clinical level was based on two
randomised trials: one study in patients with RA (Study GP13-201), and one study in patients with
advanced Follicular Lymphoma (study GP13-301). The primary objective of Study GP13-201 was
establishing bioequivalence of PK; the primary objective of Study GP13-301 was the confirmation of
therapeutic equivalence. As supportive evidence, PK-PD data were provided from a small-scaled
observational study in Japanese patients with indolent NHL.
CHMP guidelines
The following guidelines are considered of special interest:
Table 1 Guidelines
Guideline Document Reference Topic
Guideline on Similar Biological Medicinal
Products containing Biotechnology-Derived
Proteins as Active Substance: Non-Clinical
and Clinical Issues
EMEA/CHMP/BMWP/42832/2005
Rev 1, 2014
Development plan
Guideline on Similar Biological Medicinal
Products
CHMP/437/04 rev 1, 2014 Development plan
Guideline on similar biological medicinal
products containing monoclonal antibodies
– non-clinical and clinical issues
EMEA/CHMP/BMWP/403543/2010 Development plan
Guideline on the investigation of
bioequivalence
CPMP/EWP/QWP/1401/98 Rev.
1/ Corr **
PK trial design
Guideline on the clinical investigation of the
pharmacokinetics of therapeutic proteins
CHMP/EWP/89249/2004 PK trial design
Guideline on Immunogenicity Assessment
of Biotechnology-derived Therapeutic
Proteins
EMEA/CHMP/BMWP/14327/2006 PK and
efficacy/safety trial
design
Guideline on the evaluation of anticancer
medicinal products in man
EMA/CHMP/205/95/Rev. 4 Efficacy trial design
Draft Guideline on clinical investigation of
medicinal products other than NSAIDs for
treatment of rheumatoid arthritis
CPMP/EWP/556/95 Rev. 2 Efficacy trial design
Guideline on the choice of the
non-inferiority margin
EMEA/CPMP/EWP/2158/99 Efficacy trial design
Assessment report
EMA/303207/2017 Page 14/115
2.2. Quality aspects
2.2.1. Introduction
The finished product is presented as a sterile concentrate for solution for infusion containing 100 mg
(in 10 mL) or 500 mg (in 50 mL) of rituximab as active substance.
Other ingredients are: sodium citrate, polysorbate 80, sodium chloride, sodium hydroxide, hydrochloric
acid, water for injections.
The product is available in 10 mL or 50 mL clear glass vials with butyl rubber stopper containing 100 or
500 mg of rituximab. Packs of 10 mL (100 mg) contain 2 or 3 vials. Packs of 50 mL (500 mg) contain 1
or 2 vials.
The formulation of the finished product was developed to maintain the similarity to the EU-marketed
reference product MabThera.
2.2.2. Active Substance
General information
The International Nonproprietary Name (INN) of the active substance contained in Rixathon is
rituximab. It is a murine/human chimeric IgG1 kappa type monoclonal antibody directed against the
CD20 antigen found on the surface of normal and malignant B lymphocytes. This chimeric anti-CD20
antibody is produced by recombinant DNA technology in a Chinese Hamster Ovary (CHO) mammalian
cell expression system. It has the characteristics which are common in monoclonal antibodies and
include amino acid modifications such as deamidation, oxidation or glycation, disulfide bridging,
variable N-glycosylation, N- and C-terminal heterogeneity, and molecular weight variants.
Manufacture, characterisation and process controls
Description of manufacturing process and process controls
The active substance is manufactured according to current Good Manufacturing Practices (cGMP) by
Sandoz GmbH Schaftenau, Austria. The manufacturing process reflects a standard process used for the
manufacture of monoclonal antibodies. The active substance is produced in a fed-batch process. The
cell culture process involves three stages: the seed train, the inoculum train, and the production
culture. The purification process consists of consecutive operations including primary separation;
several chromatography steps; viral inactivation/filtration and UF/DF. Description of the active
substance manufacturing process is considered adequate.
Process characterization was conducted in accordance with current ICH requirements including quality
by design (QbD) principles. Process characterization included large scale and small scale data mining,
risk assessments, and laboratory studies. A process risk assessment tool based on failure mode and
effects analysis (FMEA) methodology was used to assess the process- and product-related risks of the
active substance manufacturing process and to select the process parameters (PPs) which should be
further investigated during process characterisation (PC) studies. The classification of PP is
appropriate.
Assessment report
EMA/303207/2017 Page 15/115
Control of materials
As the active substance has been developed as biosimilar, the coding sequence of the expression gene
was designed to achieve the identical primary amino acid sequence as the reference product.
Cell bank testing and characterisation was performed according to the requirements defined in ICH
Q5A and Q5D. The data of cell bank characterisation demonstrate the absence of microbial and viral
contaminants, as well as endogenously encoded retrovirus-like particles. Protocols for the preparation
and testing of future Working Cell Banks are in place to ensure consistent supply.
In the manufacturing process of the active substance no material of human or TSE relevant species is
used.
Control of critical steps and intermediates
An extensive control strategy is proposed. Process controls performed during manufacture of the active
substance are categorized.
Process validation
Process validation at commercial scale was performed and process and performance parameters have
been maintained within the specified acceptance ranges. Results of IPCs complied with the pre-defined
limits. The release data for the three consecutive process validation batches comply with the
specifications. Process-related impurities are effectively removed by the purification process. Data from
manufacturing scale and results from spiking studies at small scale show that process related
impurities are reduced to concentrations below the acceptable limit, which is based on worst case
considerations taking a maximum intravenous dose of 1,000 mg per day into account. The removal of
product related impurities was demonstrated. Overall the data show that the process is under control
and suitable for consistent manufacturing. A small scale study showed that the reuse of the
chromatography resins has no adverse effects on process performance and product quality including
product and process related impurities and adventitious agents. Based on the results of the small scale
study, the maximum number of cycles of the chromatography resins was defined. The results are
supported by concurrent manufacturing scale validation activities. Hold times were established and are
considered validated.
Manufacturing process development
In the first development phase a downstream process suitable for manufacturing of AS for preclinical
and clinical studies was successfully developed. During the second development phase the downstream
process was further improved. The comparability data demonstrate that the derived material is
comparable to the reference material.
Characterisation
For the characterisation of Rixathon a comprehensive series of analytical methods have been used.
These methods included state-of the art sensitive and orthogonal physicochemical and biological tests
to determine the primary, secondary, and higher-order structure, post-translational modifications and
associated heterogeneities, glycosylation, charge variants, purity/impurities, and quantity of Rixathon.
Molecular Mass and Primary Structure: the active substance is a 145 kDa monoclonal antibody
composed of two light chains (213 amino acid) and two heavy chains (451 amino acid), which are N-
glycosylated at Asn 301. Based on its theoretical sequence and characterisation studies, peptide
mapping confirmed that the active substance had the expected primary structure and it can be stated
Assessment report
EMA/303207/2017 Page 16/115
that the sequence of the active substance is identical to the theoretical sequence of rituximab. In
addition mass spectrometry analyses showed that all test items had the expected masses.
Disulfide bridging: All disulfide linkages could be confirmed by peptide mapping and x-ray
crystallography. The levels of free thiols were comparable in all test items.
N-Glycosylation: Analyses of oligosaccharides showed that the active substance had consistent
oligosaccharide distribution and expected glycosylation for an antibody produced in CHO cells, showing
one single N-glycosylation site at the heavy chains (Asn301). No potentially immunogenic glycoforms
such as NGNA or Gal-α1,3-Gal could be identified.
Charged variants: Charge heterogeneity was evaluated for all test items and revealed consistent
values for acidic and basic peak variants, the amount of glycated variants as well as for sialylated
structures.
Molecular size variants: Size heterogeneity was assessed by capillary gel electrophoresis, SEC, SEC-
MALLS, AUC, AF4, MFI, light obscuration and visible particles determination according to Ph. Eur. The
results show that all variants detected were of proteinous nature.
Higher order structure: Structural analyses of the active substance showed that all test items had
identical higher order structures.
Biological function: Rituximab has a number of elements that are known to contribute to its mode of
action. After binding of the CD20 antigen on the surface of B cells, the complement system is activated
via binding of C1q to the Fc part of the antibody, leading to complement dependent cytotoxicity (CDC).
In addition, the antibody can interact with FcR positive cells and can thereby induce antibody
dependent cellular cytotoxicity (ADCC). Furthermore, binding of rituximab can induce apoptosis of the
target cell. All elements of these modes of action were tested using different types of assays: CD20
binding activity (binding assay), FcRn (SPR), FcγRIa (SPR), FcγRIIa (SPR), FcγRIIb (SPR), FcγRIIIa
(F158) (SPR), FcγRIIIa (V158) (SPR), FcγRIIIb (SPR), ADCC activity (ADCC assay), CDC activity (CDC
assay, C1q binding (C1q binding assay) and apoptosis induction (apoptosis assay). All assays revealed
consistent results for the active substance tests with all values within the specified and expected
ranges.
The evaluated data confirmed the capability of the active substance manufacturing process to produce
consistent batches. The results obtained showed that the active substance has the expected primary,
secondary and tertiary structures and physicochemical properties of a human IgG1 type antibody and
all biological characteristics revealed no result being out of the specified or defined target range.
The active substance specification includes test methods for clarity, coloration, pH, identity, purity,
glycosylation profile, bioburden, endotoxins, content and biological activity (potency).
Characterisation is additionally discussed in the context of biosimilarity to the reference product.
Specification
The specifications set for the release of the active substance have been set taking ICH Q6B guideline
into account.
Clearance validation studies have been performed to demonstrate that the manufacturing process
provides adequate clearance of impurities. Process-related impurities are extensively identified and
assessed. The batch results indicate that levels of process-related impurities are consistently low
among the AS batches. Product-related impurities have been adequately defined and are also
Assessment report
EMA/303207/2017 Page 17/115
discussed as part of the biosimilarity data. Sufficient information is provided to substantiate that the
CPPs/CIPCs guarantee satisfactory impurity removal over the whole range of the associated PARs.
Analytical methods
General tests (Clarity, Colour, and pH) and Safety tests (Endotoxin and microbial enumeration) are
performed according to the Ph. Eur. monographs. Non-compendial methods are briefly described
including preparation, procedure, system suitability and assay acceptance criteria. The analytical
methods are appropriately validated.
Batch analysis
The batch results of the currently available AS batches manufactured with the proposed commercial
process at the GMP manufacturing facility at Sandoz GmbH Schaftenau, Austria indicates that the
manufacturing process is robust. All acceptance criteria were met. In order to establish the acceptance
criteria for the commercial specifications, batch data were evaluated. Statistical analysis of the batch
results has been also performed for the quantitative release tests to confirm that the proposed
commercial specifications reflect assay variability, future process variability and capability
appropriately.
Reference materials
For active substance reference material a two-tiered approach, including primary in-house reference
material and working standards, has been used. The working standards are designated to be released
compared to the primary in-house standard, which should thereby last for a longer period of time.
Stability
A comprehensive stability program has been provided. Results on long-term storage conditions have
been provided and are considered sufficient to justify the proposed shelf-life. Evaluation of the results
of the stability of the active substance at 40 ± 2°C/75 ± 5% RH showed that the selected parameters
are stability indicating and appropriate to demonstrate stability of the product. A follow up stability
program is proposed covering long-term storage conditions. The analytical methods cover stability
indicating (including potency) and safety parameters. A photo-stability study revealed only minor
effects on the purity of the active substance. Freeze-thawing and freeze-freezing studies demonstrated
the active substance stability for several cycles of alternating freezing at long-term and intermediate
frozen storage conditions.
2.2.3. Finished Medicinal Product
Description of the product and pharmaceutical development
The finished product is a sterile concentrate for solution for infusion for intravenous (IV) use after
dilution. The liquid formulation is based on rituximab as the active substance at a concentration of 10
mg/mL provided in 10 mL vials (100 mg) and 50 mL vials (500 mg). The excipients for the formulation
are compliant with the requirements of the Ph. Eur. and commonly used in parenteral medicinal
products. The applicant sufficiently described the pharmaceutical development of the finished product
(FP).
The process characterization included data mining, risk assessment, laboratory studies and technical
batches manufactured at commercial scale. A process risk assessment tool based on FMEA
Assessment report
EMA/303207/2017 Page 18/115
methodology was used to assess the process-related risks of the FP manufacturing processes and to
select the process parameters for characterization studies.
During development the manufacturing process of the FP has been changed. Initially the FP has been
produced at the first biopharmaceuticals manufacturing site. The process was afterwards transferred to
the commercial manufacturing site. Comprehensive studies have been performed to investigate the
comparability of the finished product manufactured. The results showed comparable FP as all
comparability assessment criteria have been met.
The primary container closure consist of a type I glass vial and a chlorobotyl rubber stopper. Both
components meet the requirements of the Ph. Eur. The vials are crimped with an aluminium cap with a
flip-off component. The compatibility of all constituents of the finished product with the container
closure system was sufficiently demonstrated by stability data of numerous batches.
Manufacture of the product and process controls
The finished product is manufactured according to current Good Manufacturing Practices (cGMP). It is
released by Sandoz GmbH Schaftenau, Austria... The finished product is produced using standard
manufacturing steps such as thawing of the AS, dissolving of excipients, compounding, sterile filtration
and aseptic vial filling. Sterile filtration by means of bacteria-retentive membrane filters followed by
aseptic filling is applied.
The process is sufficiently described including details on stirring, filtration and filling. Additionally, an
overview of process parameters and their target values/acceptable ranges is provided.
The established in-process controls (IPCs) are considered as appropriate tests to monitor the process
and assure a consistent performance of the manufacture of the FP. Classification of process parameters
(PPs) was performed taking into account the existing product, process knowledge, and experimental
data. Considering that the finished product manufacturing process is straightforward, the proposed
IPCs, PPs and their ranges are acceptable. Validation of the manufacturing process was executed as
prospective validation for the 100 mg and 500 mg strengths of the finished product. Process validation
data / process qualification data demonstrate that when producing within the process conditions set,
the predefined IPCs and product specifications are met.
The control strategy describes how QAs are addressed through different control elements. It provides a
link between the finished product quality and the control elements that are established to ensure
process robustness and product quality.
Subsequent to successful process validation, process performance and product quality are monitored
as part of the continued process verification (CPV) to ensure that the state of control is maintained
throughout the commercial manufacturing process. Continued process verification is a planned life-
cycle management program to ensure that the manufacturing process remains capable and is in a
state of control. This is achieved through the systematic collection, analysis and trending of product-
related and process-related data.
Product specification
The specifications for the release of the finished product have been set taking the principles of the
guideline ICH Q6B into account. The finished product specification includes test methods for
coloration, clarity, pH, extractable volume, osmolality, identity, purity, sterility, endotoxins, visible and
sub-visible particles content and biological activity (potency).
Justification is presented in support of the proposed set of the finished product release / shelf life tests
and their acceptance criteria.
Assessment report
EMA/303207/2017 Page 19/115
Analytical methods specific for the FP are briefly described. For compendial methods, the applicant
refers to the corresponding Ph. Eur. monographs. For methods (including validation) identical for the
AS testing the applicant refers to the corresponding AS sections. The suitability of compendial methods
was verified for their use. The validation of the tests was adequately completed. . Batch analyses data
is provided for laboratory scale, pilot scale and commercial scale batches, confirming the consistency of
the manufacturing process and its ability to manufacture to the intended product specification. Each
batch was tested to the specification in place at the time of manufacture and all specifications were
met. For the finished product the same reference materials are used as for the active substance.
Stability of the product
The applicant provided a comprehensive stability program including long-term, accelerated, stress
conditions as well as freeze / thaw, out-of-the fridge, inverse out-of-the fridge and photostress studies.
The stability protocols, including the selected QAs to be tested and time points for sampling, are
acceptable and in accordance with ICH Q5C. For the long-term storage conditions, data were evaluated
statistically using ANOCOVA for all stability-indicating parameters and a theoretical shelf-life according
to ICH guideline Q1E was calculated where applicable.
The stability batches encompassed both strengths 100 mg and 500 mg. Data up to 36 months at long-
term conditions have been presented for batches.
For most of the quality attributes tested the results show a highly stable finished product at the
recommended long-term stability conditions (5 ± 3°C). Importantly, all results of the tested finished
product 100 mg and 500 mg batches were within the shelf-life specifications during storage at 5 ±
3°C.
Evaluation of the results of the stability of the finished product at accelerated (25 ± 2°C / 60 ± 5%
RH) and stressed (40 ± 2°C / 75 ± 5% RH ) conditions showed highly comparable degradation
profiles.
In-use stability study
The in-use study performed with the finished product diluted in 0.9% NaCl and 5% Glucose and stored
in PE bags revealed no changes for the biophysical stabilities of the FP after in-use storage under
different storage conditions.
The results of the freeze / thaw studies did not reveal a significant impact on the quality attributes of
the FP. Furthermore, the FP has been shown to remain stable at recommended conditions for up to 36
months after initial storage at room temperature for up to 14 days (out-of-the fridge study). The
inverse out-of-the fridge study (14 days at room temperature after 36 months storage at
recommended conditions) showed slight differences in the charged variants. The results stayed within
the defined acceptance criteria. The results of the photostress study demonstrated that illumination
slightly induces degradation in the FP.
Adventitious agents
Compliance with the TSE Guideline (EMEA/410/01 – rev. 3) has been sufficiently demonstrated. The
active drug substance of Rixathon is produced in a serum-free culture medium. No TSE relevant
material is added during cell cultivation of the active substance. The MCB and WCB which have been
established are free from TSE-risk substances.
The active substance is expressed in CHO cell using serum-free medium. The cell banking system has
been extensively screened for adventitious viruses using a variety of in vitro and in vivo assays. The
tests demonstrated the absence of any virus contaminants in the cell banks with the exception of
Assessment report
EMA/303207/2017 Page 20/115
intracellular type A and extracellular type C retrovirus-like particles which are well known to be present
in rodent cells; this is acceptable as there is sufficient capacity within the active substance
manufacturing process to inactivate/remove such virus particles. . Both enveloped and non-enveloped
viruses were effectively reduced by the various unit operations including inactivation at low pH,
chromatography steps and a virus filtration step. Chromatography resins are re-used during the active
substance manufacturing process and both new and used resins have been investigated with very
similar performance with respect to virus reduction. At the end of the active substance cell culture
procedure, general testing for adventitious viruses is performed and a cell line is included to detect
minute virus of mice.
Biosimilarity
The overall strategy to demonstrate biosimilarity between Rixathon and MabThera, the EU authorised
reference product, was designed in a stepwise approach: demonstrating analytical comparability
between the biosimilar, the reference product (MabThera) and the US authorised Rituxan; and by
generating non-clinical and clinical data.
As a first step, in an extensive structural and functional characterization, analytical comparability
between the biosimilar and the reference product was established. A comprehensive set of binding and
activity assays were conducted to gain a full understanding of all functionalities of the molecule and
the structures underlying these functionalities that contribute to its MoAs, demonstrating that the
active substance and the finished product batches of the biosimilar and MabThera/Rituxan batches are
comparable on physicochemical and biological level. A sufficiently high number of the biosimilar
batches manufactured by the commercial process as well as of the MabThera reference product
authorised in the EU was included in the analytical similarity exercise.
The second step involved non-clinical testing confirming the comparability of the biosimilar and
MabThera/Rituxan.
Finally, clinical comparability was based on two pivotal studies.
In order to define a biosimilarity range for the biosimilar, the applicant comprehensively analysed both
the EU reference product and US Rituxan using orthogonal methods analysing specific quality
attributes. The results showed high comparability between both products. Different batches of the EU
reference product and Rituxan were continuously monitored to verify batch to batch consistency and
showed no differences between them. This monitoring assessed the variability of the reference product
over a period of approximately 9 years. A quality shift was observed in 2008 in for the EU reference
product and US Rituxan. Several quality attributes were affected (e.g. charge variants, glycan
structures, ADCC) by this shift but both qualities were on the market simultaneously, therefore both
quality profiles observed for the reference product are considered to represent a safe and effective
product and considered appropriate for use in comparability assessment.
The primary structure of the biosimilar is identical to the primary structure of the reference product.
Using different endoproteinases and combinations of endoproteinases with distinct substrate specificity
and subsequent LC-ESIMS and MS/MS analyses 100% coverage of the amino acid sequences of the
biosimilar and the reference product was achieved. Considering also the molecular mass analyses, the
amino acid sequences are considered identical. In addition to X-ray crystallography, native LysC
peptide mapping using RP-HPLC UV/MS revealed a similar disulphide bridge pattern in the biosimilar
and the EU reference product. Analysis of the free thiols for the biosimilar was within the range of
variability of the EU reference product. Post-translational modifications (PTM) as N-terminal
pyroglutamate, C-terminal lysine variants and proline amide were identified in the biosimilar as well as
in the EU reference product, with slight higher values for proline amide in the biosimilar. However, the
Assessment report
EMA/303207/2017 Page 21/115
small difference seen in the value of proline amide is scientifically justified to not compromise
biosimilarity. Furthermore, PTMs as methionine oxidation, asparagine deamidation with or without
subsequent isomerization were analysed by mass spectrometry using batches stored under stress
conditions and identified identical locations on the antibodies for these PTMs in both the biosimilar and
the reference product.
The higher order structure of the biosimilar and the reference product has been elucidated using
orthogonal state-of the art assays. The results show highly comparable structures in terms of the
secondary and tertiary structure conformation. Furthermore x-ray crystallographic analyses
demonstrated similar Fab- and Fc-fragments as well as similar disulfide bridges within the fragments of
the biosimilar and the reference product. Thermograms showed comparable unfolding of both mAbs
further demonstrating their biosimilarity.
The elucidation of molecular size variants showed that the biosimilar has a comparable purity to the
reference product. Size exclusion chromatography (SEC) results suggest a slightly higher purity,
whereas orthogonal methods as analytical ultracentrifugation and SEC-MALLS show comparable
purities. Furthermore, in terms of the hydrodynamic diameter or polydispersity both the biosimilar and
the reference product show similar results. In addition, the enumeration of sub-visible particles
measured by resonant mass measurement and micro-flow imaging demonstrated comparable results
considering the high variability of these methods. Taken together the data on molecular size variants
justifies a biosimilarity claim of the biosimilar to its reference product.
The pattern of charged variants is distinctive for biotechnologically manufactured mAbs. Using cation
exchange chromatography, the biosimilar pattern could be resolved into acidic, main, and basic
variants. The basic variants were identified to contain C-terminal lysine, proline amidation as well as N-
terminal glutamine. The values lie within the upper range of the reference product. Acidic variants
representing mainly molecule fragments, deamidation, glycation, and pyroglutamate are lower in the
biosimilar compared to the reference product. Further analysis of the different fractions of the CEX
chromatogram (of the biosimilar and the reference product) showed that all fractions that were
manageable to purify showed comparable potency values. The glycation, which has been shown to
impact the potency of certain mAbs, showed a lower value for the biosimilar in comparison to its
reference product. Further in depth analysis of the biosimilar and the reference product with
intentionally increased glycation values did not reveal negative effects on efficacy. As the overall
values are very low and the effects on efficacy and safety minor they are considered to not question
the biosimilarity of the biosimilar and its reference product.
The comparison of the glycosylation pattern of the biosimilar and the reference product shows minor
differences. The main glycan structures identified in rituximab are bG0, bG1 and bG2, which showed
comparable values in the biosimilar and the reference product. Afucosylated structures, as well as high
mannose sugars have been shown to influence Fc-mediated effector functions such as ADCC.
Compared to the reference product, the glycan profile cumulates to comparable ADCC activity of the
biosimilar and the reference product and, taking also the overall low content of these structures into
account, no impact on efficacy/safety is expected.
The biological functions have been directly analyzed using bioassays for ADCC, CDC and Apoptosis
induction or by measuring the binding of the Fc-part to the responsible receptors Fcγ-receptors for
ADCC and C1q for CDC. A comparative stability study was performed for the biosimilar, the EU
reference product, and US marketed Rituxan, to assess the stability under long-term, accelerated and
stress conditions. The results of the studies show no significant differences in the degradation profiles
of the biosimilar and the reference product.
Assessment report
EMA/303207/2017 Page 22/115
During the comparability exercise small differences (high mannose structures, differences in charged
variants) have been found between the biosimilar and the reference product. To provide further
assurance that these differences in high mannose structures and charged variants do not affect
biosimilarity or potency, in-depth analyses as well as forced degradation studies have been performed.
The biosimilar and the reference product have been stored under stress conditions to foster
degradation, and subsequently fractionated. The potency of these fractions has been analysed and a
comparison between the biosimilar and the reference product was made. The results demonstrated
that the biosimilar and the reference product behave similarly under stressed conditions, including the
potency of the fractions. Fragments of the biosimilar have been demonstrated, as expected, to be
biologically inactive and are defined as impurities. It was demonstrated that the amounts of fragments
in the biosimilar and the reference product are comparable. Furthermore, process-related impurities as
host cell DNA and protein that affect the safety evaluation of the medicinal products have been found
to be comparably low for both products.
Summarising, a sufficiently high number of the biosimilar batches manufactured by the commercial
process as well as of the MabThera reference product authorised in the EU was included in the
analytical similarity exercise. As the US authorised Rituxan has been used in addition to the reference
product for pre-clinical studies, investigation of Rituxan batches were also included in the analytical
similarity exercise and it can be concluded that bridging between the US comparator and the EU
reference product is acceptable.
Based on the comprehensive analytical comparability exercise, similarity between Rixathon and the
reference medicinal product is considered demonstrated on quality level.
2.2.4. Discussion on chemical, pharmaceutical and biological aspects
The applicant provided a well-structured quality dossier. In general, high quality scientific data have
been presented.
The manufacturing process reflects a standard process used for the manufacture of monoclonal
antibodies and is well described, characterized and validated. The CQA assessment based on the in-
house risk assessment tool is considered acceptable. The manufacturing process and associated control
strategy are based on a structured process development strategy; ICH Q8 elements (e.g. risk
assessment tools for identification of CQAs; systematic definition of a control strategy) have been
appropriately incorporated. The data package comprises both full scale and small scale process
characterisation and validation studies; in addition full scale PPQ (verification) batches are available;
column/resin life time studies are addressed using continuous process verification elements. For the
characterization, a comprehensive series of analytical methods have been used. These methods
included state-of the art sensitive and orthogonal physicochemical and biological tests to determine the
primary, secondary, and higher-order structure, post-translational modifications (PTMs) and associated
heterogeneities, glycosylation, charge variants, purity/impurities, and quantity of Rixathon. The control
of the AS and FP is considered sufficient.
The TSE virus safety of the finished product has been sufficiently demonstrated. Rixathon has been
developed as a similar biological medicinal product to the European Union (EU)-authorised reference
product MabThera (rituximab). Overall the analytical comparability data suggest that Rixathon can be
considered biosimilar to the reference product. The applicant performed an extensive and structured
comparability exercise in order to demonstrate analytical comparability and justify biosimilarity. This
comparability exercise is supported by a risk based CQA assessment, in order to rank the attributes
and assess the impact if ranges do not overlap. The comparability exercise includes different batches
Assessment report
EMA/303207/2017 Page 23/115
of reference product, and batches of the biosimilar. Most tests directly support analytical comparability,
because ranges of the biosimilar are within the ranges for the reference product. In a number of cases,
it is justified that ranges do not overlap because the biosimilar contains less product-related
substances/impurities (e.g. aggregates).
In summary, from a quality perspective, it is considered that similarity between Rixathon and the
reference product was shown. Minor differences were identified, which are not considered to impact
efficacy and safety of the product nor preclude biosimilarity.
2.2.5. Conclusions on the chemical, pharmaceutical and biological aspects
The quality of this product is considered to be acceptable when used in accordance with the conditions
defined in the SmPC. Physicochemical and biological aspects relevant to the uniform clinical
performance of the product have been investigated and are controlled in a satisfactory way. Data has
been presented to give reassurance on viral/TSE safety. Based on the comprehensive analytical
comparability exercise, similarity between Rixathon and the reference medicinal product is considered
demonstrated on quality level. As the US authorised Rituxan has been used in addition to the reference
product for pre-clinical studies, investigation of Rituxan batches were also included in the analytical
similarity exercise and it can be concluded that bridging between the US comparator and the EU
reference product is acceptable.
2.2.6. Recommendation(s) for future quality development
In the context of the obligation of the MAHs to take due account of technical and scientific progress,
the CHMP recommends the following points for investigation:
It is recommended that the active substance and finished product specifications will be re-evaluated
after an appropriate and agreed number of batches becomes available using statistical tools, as
already committed by the applicant.
2.3. Non-clinical aspects
2.3.1. Introduction
Non-clinical animal studies included a series of studies to characterize and compare the non-clinical
pharmacodynamics (PD), pharmacokinetics (PK), and safety profiles of Rixathon and MabThera
including two pivotal GLP studies. The Applicant also submitted data from studies in two xenografted
tumour models in mice.
2.3.2. Pharmacology
Primary pharmacodynamic studies
Rituximab binds specifically to the transmembrane antigen CD20, a non-glycosylated phosphoprotein
expressed on the surface of pre-B and mature B lymphocytes, but not on hematopoietic stem cells and
terminally differentiated antibody-producing plasma cells, or other tissues. While the Fab domain of
rituximab binds to the CD20 antigen on B lymphocytes, the Fc domain can recruit immune effector
Assessment report
EMA/303207/2017 Page 24/115
functions to mediate B cell lysis (resulting in B cell depletion), as well as modulating exposure. The
ascribed mechanisms of effector-mediated cell lysis include antibody-dependent cellular cytotoxicity
(ADCC); Complement-dependent cytotoxicity (CDC); Apoptosis and Macrophage-mediated antibody-
dependent cellular phagocytosis (ADCP) – Figure 1.
Rituximab coated B cells are killed by at least 4 different mechanisms. (A) Binding of rituximab to CD20 on
B cell surface causes activation of the complement cascade, which generates the membrane attack complex (MAC) that can directly induce B-cell lysis by complement-mediated cytotoxicity (CDC). (B) Binding of rituximab allows interaction with NK cells via Fc receptors III (FcRIII), which leads to antibody-dependent cell-mediated cytotoxicity (ADCC). (C) The Fc portion of rituximab and the deposited complement fragments allow for recognition by both FcR and complement receptors on macrophages, which lead to phagocytosis and ADCC. (D) The crosslinking of several molecules of rituximab and CD20 in the lipid raft determine the interaction of these complexes with elements of a signaling pathway involving
Src kinases that mediate direct apoptosis (From Jaglowski et al 2010).
Figure 1 - Mechanisms of rituximab-mediated cell death.
In vitro pharmacodynamic studies
In an ex vivo whole blood depletion assay (study GP13-021), collected human whole blood is incubated
ex vivo with different concentrations of Rixathon. After incubation, during which concentration-
dependent B cell depletion occurs, distinct subsets of blood cells are stained with fluorescence-labelled
detection antibodies and analysed on a flow cytometer. The concentration-dependent decrease of B
cells within the lymphocyte populations is determined. Relative B cell depletion (BCD) is calculated. In
this assay, Rixathon was shown to be similar to Rituxan and MabThera in its ability to deplete B cells,
with all tested batch samples of Rixathon displaying similar biological activity at equivalent
concentrations.
Assessment report
EMA/303207/2017 Page 25/115
Study GP13-017 compared Rixathon and MabThera for in vitro ADCC potency of freshly-isolated human
peripheral blood NK cells as the effector cells against different immortalized B cell lines and the overall
results indicated that Rixathon and MabThera were similar in their ability to mediate ADCC in these cell
lines. The results from study GP13-022, which was an extension of GP13-017 using one B cell line,
indicated a comparable range of EC50 values (relative to a reference standard) for MabThera and
Rixathon.
The Applicant also assessed the depletion of a B cell line by Rixathon, MabThera, and Rituxan mediated
by freshly isolated human PBMC. This dataset, albeit limited, confirmed comparable activity of Rixathon
and the reference medicinal product.
In vivo pharmacodynamic studies
The pharmacodynamic effects of Rixathon and MabThera were compared in Cynomolgus monkeys.
After single dose iv administration of 5 mg/kg, the AUEC ratio was similar for the first 7 days, but
when the whole 28 days observation period was compared the effect of Rixathon was slightly less than
of Mabthera (AUEC ratio 0.92; 95% CI 0.87-0.98) for the CD20low B-cells, which are considered more
close to human CD20 cells. For the CD20high B-cells, the AUEC ratio of 1 was within the 95% CI (AUEC
ratio 0.94; 95% CI 0.88-1.01).
In the repeated dose toxicology study the same PD parameters were evaluated after 4 weekly iv
administrations of 20 or 100 mg/kg. With these dose regimens no significant differences were
observed.
In study GP13-015, a slight, but non-significant difference in survival was observed in SCID mice
injected with Granta-519 cells from human mantle cell lymphoma cell line treated with 40 mg/kg
Rixathon (group 3; mean survival ± S.D.: 34.9 ± 3.74 days) and 40 mg/kg Mabthera (group 5; mean
survival ± S.D.: 36.1 ± 2.63 days).
In study GP13-018, where SCID mice were injected with Raji human Burkitt lymphoma cells, 1.25
mg/kg Rixathon was significantly more effective than 1.25 mg/kg MabThera (P = 0.0485). Differences
between the agents at the lower and higher dose levels were non-significant and overall no dose-
dependency was observed.
Assessment report
EMA/303207/2017 Page 26/115
Figure 2 - Study GP13-011 Mean Tumour Growth in SU-DHL-4-model: In vivo comparability of Rixathon and MabThera
Dose-response effects of Rixathon and MabThera in human mantle cell lymphoma Jeko-1 xenograft tumours.
Female SCID beige mice were implanted subcutaneously with Jeko-1 cells into the right flank. When tumours
reached a certain size, mice were dosed with either Privigen (IgG control antibody) or Rixathon or MabThera at
indicated doses. Arrows indicate dosing days.
Figure 3 - Study GP13-014 Antitumour activity of Rixathon, MabThera, and Privigen (IgG control antibody) against subcutaneous Jeko-1 xenograft tumours
Assessment report
EMA/303207/2017 Page 27/115
Data from both Study GP13-011 and Study GP13-014 indicate that treatment with Rixathon or
MabThera at sub-therapeutic dose levels results in comparable tumour growth inhibition in mouse
xenograft models of NHL. The relative anti-tumour activity of Rixathon and MabThera remained
comparable throughout the observation periods and at all tested dose levels in both studies, despite
the increase in intra-group heterogeneity at later time points that is characteristic of these xenograft
models.
Secondary Pharmacodynamics
Dedicated secondary PD studies have not been submitted (see discussion on non-clinical aspects).
There were no reports of toxicity caused by a lack of specificity for the primary target (i.e. there are no
known off-target effects of rituximab) and there was no off-target binding of Rixathon in an in vitro
cross-reactivity study performed in a comprehensive panel of human tissues.
Safety Pharmacology
Dedicated safety pharmacology studies have not been submitted (see discussion on non-clinical
aspects).
2.3.3. Pharmacokinetics
Pharmacokinetic/toxicokinetic response was evaluated in Cynomolgus monkeys after a single 5 mg/kg
dose for 9 days and after two weekly doses of 20 or 100 mg/kg for 14 days.
Figure 4 - Study GP13-007: Geometric mean serum concentration of rituximab in male cynomolgus monkeys over nine days after a single intravenous dose of Rixathon or MabThera (5 mg/kg)
Assessment report
EMA/303207/2017 Page 28/115
Table 2 - Study GP13-007: Pharmacokinetic parameters in male cynomolgus monkeys after a single intravenous administration of Rixathon or MabThera (5 mg/kg)*
Table 3 - Study GP13-008: Ratio of area under the serum concentration-time curves and of maximum serum concentration in cynomolgus monkeys after once-weekly repeated intravenous doses of Rixathon or MabThera
2.3.4. Toxicology
Single dose toxicity
No dedicated single-dose toxicity studies were performed for Rixathon.
In a single dose PK/PD study in male cynomolgus monkeys a dose of 5 mg/kg of each product was
administered i.v. to 14 animals per group. Both test items showed comparable safety profiles
consistent with the pharmacology of rituximab when evaluated for up to ten weeks following dose
administration.
Assessment report
EMA/303207/2017 Page 29/115
Repeat dose toxicity
Study GP13-008 Rixathon / MabThera: Comparative repeated dose toxicity study in cynomolgus monkeys
This was a 4 week study with the purpose to compare the safety profile of Rixathon (commercial scale
quality), to that of MabThera, following repeated iv administration to cynomolgus monkeys on the days
1, 8, 15, and 22 of the study period plus a 4-week dosing-free period, and, in a subset of the animals,
to assess comparability of the reversibility of the effects observed during a 6 month recovery phase.
There were no detectable differences between test and reference item. Neither substance induced any
toxicologically relevant findings in weekly dosing of 20 or 100 mg/kg for 4 weeks. Pharmacological
findings were a massive but not total reduction of B lymphocytes starting on day 2 of dosing, which
resulted in shifted immunophenotyping values and correlated with lacking germinal centres in
lymphatic organs, accessory in single animals a lymphoid depletion of spleen or axillary lymph node.
The recovery of B cell reduction was seen in all groups and comparable in test and reference item
dosed animals. The no observed adverse effect level (NOAEL) can be defined at the high dose of 100
mg/kg for the test item.
Immunogenicity evaluations were included in the comparative single- and repeat-dose studies with
Rixathon and MabThera. In both studies, cynomolgus monkey serum samples were analysed for the
presence of anti-rituximab antibodies using an ELISA method.
In the single-dose study all animals developed anti-rituximab antibodies between nine to 14 days after
treatment with either Rixathon or MabThera, each of which was evaluated at a dose of 5 mg/kg.
In the comparative repeat-dose toxicity study the number of animals that developed anti-rituximab
antibodies were comparable at equivalent dose levels of Rixathon and MabThera. Anti-rituximab
antibodies were detected in the 20 mg/kg Rixathon and MabThera groups starting on Day 15 (or 14
days after the first dose); as a result, TK analyses were only performed up to Day 15. Antibodies were
directed against the murine F(ab’)2 fragment of the drug and/or the Fc fragment of the drug. In the
100 mg/kg treatment groups, most of the animals showed no (or only a marginal) immune response,
which was likely a result of rapid and marked depletion of B cells and/or a high drug concentration-
induced tolerance.
Genotoxicity
Dedicated genotoxicity studies have not been submitted (see discussion on non-clinical aspects).
Carcinogenicity
Dedicated carcinogenicity studies have not been submitted (see discussion on non-clinical aspects).
Reproduction Toxicity
Dedicated reproductive and developmental toxicity studies have not been submitted (see discussion on
non-clinical aspects).
Assessment report
EMA/303207/2017 Page 30/115
Toxicokinetic data
Toxicokinetic data have not been submitted.
Local Tolerance
No injection site findings were noted in the repeat-dose i.v. toxicity study in which GP2013 or
MabThera were administered to monkeys once-weekly for four weeks.
Other toxicity studies
No other toxicity studies were submitted.
2.3.5. Ecotoxicity/environmental risk assessment
N/A – see discussion on non-clinical aspects
2.3.6. Discussion on non-clinical aspects
According to Guideline on similar biological medicinal products containing biotechnology-derived
proteins as active substance: non-clinical and clinical issues Guideline on similar biological medicinal
products containing biotechnology-derived proteins as active substance: non-clinical and clinical issues
EMEA/CHMP/BMWP/42832/2005 Rev1, the applicant used a stepwise approach in order to demonstrate
the Rixathon is comparable to MabThera with respect to PD/PK and toxicity.
Studies regarding secondary pharmacology, safety pharmacology, reproduction toxicology, and
carcinogenicity and on local tolerance are not required for non-clinical testing of biosimilars.
A comprehensive program of studies provided extensive data that indicate an overall comparability
between Rixathon and MabThera. Physicochemical assays have indicated that Rixathon is comparable
to MabThera/Rituxan with regard to primary and higher order structure, post-translational
modifications, and size variants. Functional characterisation included CD20 binding, C1q binding, and
surface plasmon resonance (SPR) Fc receptor affinity assays complemented by in vitro cell-based
bioassays (ADCC assay, CDC assay, apoptosis assay).
Additional whole blood and ADCC potency in vitro non-clinical studies were conducted and B cell
depletion and ADCC potency of Rixathon was compared to both MabThera and Rituxan. Comparable B-
cell depletion was demonstrated. Different ADCC assay formats were used and the overall data indicate
that EC50 values are similar.
The ADCP assay is a Reporter Gene Assay (RGA) in which FcγRIIa-mediated activation of nuclear factor
of activated T-cells (NFAT) is measured through the luciferase gene expression. The FcγRIIa-receptor
is considered an important mediator in eliciting ADCP when phagocytes interact with antibody-
opsonised target cells. Other Fcγ-receptors are expressed as well by macrophages. However their
contribution to ADCP is less certain. Furthermore, the functional activity of FcγRIIIa has already been
captured by the comparative ADCC assays performed by the Applicant. The ADCP assay developed by
the Applicant is considered appropriate to evaluate and compare the ADCP activity of Rixathon,
Mabthera and Rituxan. The results of the ADCP assay support the conclusion of biosimilarity.
Data submitted in support of this application are related to the comparison between the biosimilar and
the US rituximab (Rituxan); this approach was acceptable and relevant for the EU since the bridge
Assessment report
EMA/303207/2017 Page 31/115
between the US and EU product has been sufficiently justified from analytical studies, structural and
functional data (see discussion on quality).
The PD profiles of Rixathon and MabThera were compared in a single-dose i.v. PK/PD and in a four-
week repeat-dose i.v. toxicity study in healthy Cynomolgus monkeys.
The comparison of Rixathon and MabThera over 8 weeks revealed comparable AUECs for Rixathon and
for MabThera for both B cell subpopulations with ratios around 1.0 for both B cell subpopulations. With
respect to total exposure to rituximab, the results of the non-compartmental PK analysis confirmed
bioequivalence between Rixathon and MabThera with similar AUCs and 90% CIs lying entirely within
the standard bioequivalence acceptance range of 0.8 to 1.25.
The Applicant also included data from studies in two xenografted tumour models in mice. In these two
different mouse xenograft models of human lymphomas both Rixathon and MabThera showed positive
effects on animal survival compared to animals in the control groups. Animal survival was comparable
between Rixathon and MabThera and clinical signs and body weight changes were comparable among
all Rixathon and MabThera groups in both studies. Data from both studies demonstrate that treatment
with Rixathon or MabThera results in comparable tumour growth inhibition in mouse xenograft models
of NHL. In the SU-DHL-4 human B cell lymphoma CB17 SCID mouse xenograft model Rixathon
appeared to be slightly more active than MabThera during the dosing period. But the overall outcome
shows that treatment with Rixathon or MabThera showed no significant differences in short-term and
overall efficacy results, early tumour growth and progression, and relative anti-tumour efficacies.
Pharmacokinetic parameters were evaluated in the single dose study in monkeys and after once-
weekly i.v. (bolus) administrations of Rixathon or MabThera to male and female monkeys at dose
levels of 20 mg/kg or 100 mg/kg given over a four-week period. A competitive ELISA method for the
bioanalysis of rituximab in cynomolgus monkey serum was validated. A lower limit of quantification
(LLOQ) was established. The validated ELISA method is acceptable. The results show comparable PK
profiles at equivalent dose levels when evaluated for up to 14 days following the initial dose
administration. Lower Cmax values (by 6 to 20%) were observed for Rixathon compared to MabThera
at both dose levels which were attributed to intrinsic heterogeneity among individual monkeys and
variations in the initial sampling time point. The justification is considered sufficient since the analysis
showed comparable AUC values for the two treatment groups at equivalent dose levels, with 90% CI
ratios within the standard acceptance range of 0.8 to 1.25. Furthermore B cell depletion seems not to
be impacted by the lower Cmax.
There were no signs of toxicity and no detectable differences between groups administered equivalent
dose levels of Rixathon and MabThera in the four week repeat dose i.v. toxicity study in monkeys. The
no observed adverse-effect level (NOAEL) for Rixathon in the four week study was reported to be 100
mg/kg, the highest dose level evaluated and which exceeds clinical dose levels of MabThera/Rituxan.
There was no off-target binding of Rixathon in an in vitro cross-reactivity study performed in a
comprehensive panel of human tissues. All binding was related to the expected pharmacology of
Rixathon and its ability to bind CD20 expressing B lymphocytes, consistent with the reversible effects
that were observed on B cell numbers in the single and repeat dose studies in monkeys and with the
beneficial effects of Rixathon in mouse xenograft human B cell models of NHL.
Assessment report
EMA/303207/2017 Page 32/115
2.3.7. Conclusion on the non-clinical aspects
In the light of the overall in vitro and in vivo non-clinical data, Rixathon (rituximab) can be considered
similar to the reference product Mabthera. The non-clinical information under section 5.3 of the
Mabthera SmPC applies also to Rixathon.
2.4. Clinical aspects
2.4.1. Introduction
GCP
The Clinical trials were performed in accordance with GCP as claimed by the applicant.
The applicant has provided a statement to the effect that clinical trials conducted outside the
community were carried out in accordance with the ethical standards of Directive 2001/20/EC.
Table 4 - Tabular overview of clinical studies
Study No. Study title Study population Treatment duration
Dosage and [batch numbers]
GP13-201 (Part I-comparing GP2013 vs. MabThera- included in the dossier)
1
(pivotal)
A randomized, double-blind, controlled study to evaluate PK, PD, safety and efficacy of GP2013 and rituximab in patients with active RA refractory or intolerant to standard DMARDs and one or up to three anti-TNF therapies
Patients with active RA
Total: N = 173 (149f, 24m)
GP2013: N = 86 (76f, 10m)
MabThera: N = 87 (73f, 14m)
Age mean (range) = 53.71 (21-82) years
52 weeks plus up to 26-weeks after first infusion of second treatment, if necessary
GP2013 or MabThera: 1000 mg (10 mg/L in 500 mg (50 mL) single use vials), two single iv infusions, two weeks apart in combination with MTX
Optional 2nd
course of GP2013 or MabThera after week 24 (2x 1000 mg i.v., 2 weeks apart)
GP13-301 (data until cut-off 10 Jul 2015 included in the dossier; (pivotal)
A randomized, controlled, double-blind Phase III trial to compare the efficacy, safety and pharmacokinetics of GP2013 plus cyclophosphamide, vincristine, prednisone vs. MabThera plus cyclophosphamide, vincristine, prednisone, followed by GP2013 or MabThera maintenance therapy in patients with previously untreated, advanced stage follicular lymphoma (ASSIST-FL trial)
Patients with previously untreated, advanced stage FL
Combination phase: Total: 627 (2 of the 629 randomized patient were not treated);
Maintenance phase: Total: 462;
N (f/m) = 350/277
Age mean (range) = 56.9 (23–84) years
Up to 3 years (Combination Treatment Period: 6 months, Maintenance Treatment Period and/or Follow-up Period: 2.5 years).
Combination Treatment Period: Total 8 cycles of GP2013 or MabThera 375 mg/m² (10 mg/L in 500 mg (50 mL) single use vials) i.v. on Day 1 of each cycle +CVP administered every 21 days (±3 days)
Maintenance Treatment Period (patients with PR or CR after 8 cycles of combination treatment): 8 treatment cycles - GP2013 or MabThera 375 mg/m² i.v. administered every 3 months (±14 days) for a further 2 years
2.
GP13-101 (supportive)
Phase I trial to assess the safety and pharmacokinetics of GP2013 monotherapy administered weekly in Japanese patients with CD20 positive low tumor burden indolent B-cell non-Hodgkin’s lymphoma
Japanese patients with CD20 Positive low tumor burden indolent B-cell NHL
Total: 6;
N (f/m) = 4/2
Age mean (range) = 59 (41-69) years
12 weeks (Treatment Period: 8 weeks, Follow-up Period: 30 days)
GP2013: 375 mg/m² (10 mg/L in 500 mg (50 mL) single use vials), single i.v. infusions on Day 1 of each week for up to 8 weeks
CR = Complete response; CVP = cyclophosphamide, vincristine, prednisone; DMARDs = Disease modifying anti-rheumatic drugs; EU: European Union; f = female; FL = Follicular lymphoma; i.v. = intravenous(ly); m = male; MTX =
Assessment report
EMA/303207/2017 Page 33/115
Methotrexate; N = number of patients or subjects; NHL = non-Hodgkin’s lymphoma; PK = Pharmacokinetics; PD = Pharmacodynamics; PR = Partial response; RA = Rheumatoid arthritis; TNF = Tumor necrosis factor 1
Part II of GP13-201 study (comparison of GP2013 with Rituxan) is ongoing.
2 For Italy only, consists of 12 treatment cycles of single agent GP2013/MabThera administered every 2
months (± 14 days) for a further 2 years.
2.4.2. Pharmacokinetics
Three studies were submitted to support the pharmacokinetics part of the application: the pivotal
study GP13-201-Part 1 compared the PK profiles of Rixathon and MabThera in patients with
rheumatoid arthritis (RA), the supportive study GP13-301 compared the PK profiles of Rixathon and
MabThera in a subset of patients with follicular lymphoma (FL) and the other supportive study GP13-
101 compared the PK profiles of Rixathon and Rituxan studied in 6 Japanese patients with indolent
Non-Hodgkin’s lymphoma (NHL).
Analytical methods
ELISA was used for the determination of rituximab concentrations in human RA and FL serum. For the
determination of immunogenicity an affinity capture elution (ACE) ELISA for detection of antibodies
against rituximab in human RA serum and an electrochemiluminescence (ECL) bridging
immunogenicity assay for detection of antibodies against rituximab in human FL serum were used. In
order to determine neutralizing antibodies a cell-based complement-dependent cytotoxicity (CDC)
assay for detection of neutralizing antibody (NAb) in RA serum and a cell-based competitive ligand
binding assay (CLB) assay for detection of NAb in FL serum. The assay methods were validated
separately using the respective patient-specific serum matrix (i.e. serum of RA and FL patients) as
serum of patients suffering from either RA or FL can differ in their matrix composition.
Pharmacokinetics in target population
Pivotal PK/PD bioequivalence study GP13-201 in patients with RA (GP13-201 Study Part I)
The purpose of this clinical PK/PD study was to assess bioequivalence between Rixathon and EU-
approved MabThera as well as US-licensed Rituxan in patients with active RA. Therefore the study was
set up in two parts, Part I for assessing bioequivalence between Rixathon and MabThera and Part II for
assessing bioequivalence between Rixathon and Rituxan. Study Part I is completed; an overview of the
study design for Study Part I is shown in the figure below.
Assessment report
EMA/303207/2017 Page 34/115
n=number of patients randomized; Wk=Week
Figure 5 - Study design for GP13-201 Part 1
The study included male and female patients ≥ 18 years of age with a ≥ 6 months’ diagnosis of RA
based on the American College of Rheumatology (ACR) 1987 criteria. Patients had to be seropositive
for rheumatoid factor (RF) and/or anti-CCP antibodies, had an inadequate response or intolerance to
non-biologic DMARDs and one or up to three TNF antagonists, and had been receiving MTX (7.5 mg to
a maximum of 25 mg per week) for at least 4 months; stable dose for 4 weeks prior to randomization.
The eligible patients were randomized (1:1) to receive the first course of study medication (a 1000 mg
i.v. infusion of Rixathon or MabThera on two separate occasions, two weeks apart (i.e., on Day 1 and
on Day 15)). The duration of the first study drug infusion (Day 1) was approximately 4 h 15 min and
the duration of the second study drug infusion (Day 15) was approximately 3 h 15 min (if the first
infusion was uneventful). After the first treatment with study medication on Day 1, patients were
followed for 52 weeks. Samples for comparative PK and PD assessments were collected until Week 24
and Week 52, respectively. Efficacy and safety data were assessed on a regular basis until Week 52
and were used to evaluate the similarity of the efficacy and safety between Rixathon and MabThera.
Results
A total of 302 patients were screened, of which 173 patients met the eligibility criteria and were
randomized to either Rixathon (n=86) or MabThera (n=87) treatment group. The majority of
randomized patients (n=142, 82.1%) completed the study up to 52 weeks and few patients (n=31,
17.9%) discontinued the study. The most common reasons for premature discontinuation across both
Assessment report
EMA/303207/2017 Page 35/115
treatment groups were adverse events (n=8, 4.6%), unsatisfactory therapeutic effect (n=8, 4.6%)
and withdrawal of consent (n=6, 3.5%).
All patients in the Full Analysis Set (FAS = 86 patients for study drug, 87 patients for Mabthera)
received study drug at least once and are therefore also included in the Safety (SAF) Analysis Set.
Patients with major protocol deviations such as C-reactive protein (CRP) and Erythrocyte
sedimentation rate (ESR) values who did not meet the criteria either at baseline or at the unscheduled
reassessment visit, patients having negative RF and negative ACPA at the screening visit or patients
not been treated with randomized treatment, were excluded from the PK or Per Protocol (PP = 85 for
study drug, 82 patients for Mabthera) or both analysis sets. Pharmacokinetic analysis Set (PAS; 86
patients for study drug, 86 patients for Mabthera) was used for the analyses of PK, PD parameters.
Baseline and disease characteristics
Table 5 - Demographics by treatment (Full Analysis Set)
Assessment report
EMA/303207/2017 Page 36/115
Table 6 - Baseline disease history by treatment (Full Analysis Set)
All patients received anti-TNFs and non-biologic DMARDs prior to entering the study. The most
commonly reported prior RA-related medications (excluding anti- TNFs) were selective
immunosuppressants (23.1%) and glucocorticoids (19.1%). Etanercept was the most frequently used
(34.1%) prior TNF-α inhibitor, followed by adalimumab (23.1%) and infliximab (15.6%). There were
no relevant differences between the treatment groups in the type or frequency of use of prior
(discontinued) RA-related medications.
Assessment report
EMA/303207/2017 Page 37/115
There were no relevant differences between the two treatment groups in terms of non RA related
medications and significant non-drug therapies used (and discontinued) prior to start of study drug
(Rixathon 10.5% and MabThera 10.3%).
One patient in the MabThera group was accidentally un-blinded by a study nurse via the IRT system on
Day 32. The accidental un-blinding was recorded as a protocol deviation and the patient discontinued
from the study. The patient was not excluded from any analysis set.
Table 7 - Baseline disease characteristics by treatment (Full Analysis Set)
The mean DAS28 (CRP) scores at baseline were 5.81 (SD=0.916) and 5.85 (SD=0.880), in the
Rixathon and MabThera groups, respectively. The proportions of patients with positive anti-CCP
antibodies (ACPA) and/or rheumatoid factor (RF) were similar, approximately 98% in both treatment
arms. The mean steroid and methotrexate (MTX) doses taken were nearly identical between the
treatment arms at baseline.
Assessment report
EMA/303207/2017 Page 38/115
Primary objective (GP13-201 Study Part I)
The primary PK endpoint was AUC(0-inf) in serum samples, collected over 24 weeks. The ratio of the
geometric means (Rixathon /MabThera) of AUC(0-inf) of serum concentration up to Week 24 was
1.064 and the 90% CI [0.968, 1.169] which was within the standard bioequivalence limits.
Table 8 - AUC(0-inf) of serum concentration-time profile: Comparison between Rixathon and MabThera using ANOVA - GP13-201 Study Part I (PAS)
A sensitivity analysis including body surface area as an additional covariate provided similar results. The ratio of the geometric means (Rixathon /MabThera) was 1.054. The corresponding 90% CI (0.965, 1.151) was within the standard bioequivalence limits.
Following the first treatment course of Rixathon and MabThera, the overall exposure, AUC(0-inf), was
comparable between the Rixathon and MabThera treatment arms and the mean serum concentration-
time profiles over 24 weeks were nearly superimposable.
Figure 6 - Arithmetic mean (SD) serum PK concentration-time profile over 24 weeks by treatment (PK Analysis Set)
Assessment report
EMA/303207/2017 Page 39/115
Summary statistics for primary PK parameter (AUC(0-inf)) by treatment are provided in Table 9.
Table 9 - Summary of primary PK parameter by treatment - GP13-201 Study Part I (PAS)
A number of patients were excluded from the analyses: for AUC0-inf (GP2013: n=11/86 (12.8%),
MabThera: 16/87 (18.4%)) and for AUEC0-14d (GP2013: n=14/86 (16.3%), MabThera: 11/87
(12.6%)). The most common reason of exclusion of patients was positive ADA prior to or up to Week
24 and missing data noted to the extent that the parameter could not be derived. A number of patients
were excluded from the analyses for Cmax1 (GP2013: n=7/86 (8.1%), MabThera: 9/87 (10.3%)). The
most common reason for exclusion of patients from the Cmax1 analysis was that the post-infusion
sample was not taken within 15 minutes of end of infusion, positive ADA at pretreatment and infusion
interruption greater than 1 hr during the first infusion.
Secondary PK results (GP13-201 Study Part I)
Key Secondary endpoint Cmax1
The key secondary PK endpoint was a comparison of the maximum serum concentrations of Rixathon
and MabThera after the 1st infusion (Cmax1). Bioequivalence was defined and analysed similar to the
primary endpoint (AUC(0-inf)). The ratio of the geometric means (Rixathon /MabThera) was 1.133 and
the corresponding 90% CI was [1.017, 1.262] with its upper limit slightly outside the standard
bioequivalence limits [0.8, 1.25] (Table 10). Cmax2 was within the standard bioequivalence limits [0.8,
1.25].
Assessment report
EMA/303207/2017 Page 40/115
Table 10 - Serum Cmax (first infusion): Comparison between Rixathon and MabThera using ANOVA - GP13-201 Study Part I (PK Analysis Set)
When studying study drug administration and compliance, the two treatment arms behaved similarly
(e.g. regarding number of infusions, duration, volume of infusion.) regarding interruptions, more
patients in both treatment arms experienced infusion interruptions during the first infusion (total of 26
(15.0%) patients) as compared to the second infusion (total of 12 (7.1%) patients). When comparing
both treatment arms, there were more infusion interruptions in the MabThera arm (17.2% versus
12.9% for the first infusion of the first course) and 9.1% versus 4.7% for the second infusion of the
first course, respectively. Similar observation was made for the second course.
Other secondary endpoints
The bioequivalence criteria were also met for all secondary AUC parameters (AUC(0-14d), AUC(0-
12w), and AUC(0-24w) and Tmax (for both infusions, i.e. Tmax1 and Tmax2) as shown in Table 11.
Table 11 - Additional PK parameters (AUCs, Cmax and Tmax): Comparison between Rixathon and MabThera using ANOVA - GP13-201 Study Part I (PAS)
Assessment report
EMA/303207/2017 Page 41/115
PK/PD - Pivotal clinical efficacy and safety study in patients with FL (study GP13-301)
This was a randomised, double-blind, active-controlled, multicenter, parallel group study to compare
the efficacy, safety, PK and PD of Rixathon /MabThera in combination with cyclophosphamide,
vincristine, prednisone (CVP), followed by Rixathon /MabThera maintenance therapy in patients with
previously untreated, advanced stage FL. The study comprised four periods: Screening (up to 28 days
prior to randomization), Combination Treatment (8 cycles; approximately 6 months), Maintenance
Treatment (2 years) and Follow up (3 years from the date of randomization) (see Figure 7 below).
Figure 7 - treatment scheme in study GP13-301
After fulfilling the eligibility criteria evaluated during the screening period, the patients were
randomised in a 1:1 ratio to receive 8 cycles of either Rixathon with CVP or MabThera with CVP in the
combination treatment phase of the study (Table 12).
Table 12 - Combination treatment arms - study GP13-301
Randomisation into the two treatment arms (Rixathon -CVP or MabThera-CVP) was stratified by
Follicular Lymphoma International Prognostic Index (FLIPI) risk group (Solal-Celigny et al 2004) for
Assessment report
EMA/303207/2017 Page 42/115
low/intermediate risk (FLIPI score 0-2) vs. high risk (FLIPI score 3-5), geographic region (Asia Pacific,
Latin America, Japan and Europe) and by participating cohort for PK data collection (Cohort 2:
extensive PK sampling; Cohort 1 exclusive of Cohort 2: sparse PK sampling; rest of patients without PK
sampling). An overview of PK/PD cohorts is diagrammatically presented in Figure 8.
Figure 8 - Overview of PK/PD cohorts in GP13-301 study
196 patients were allocated to Cohort 1, in which sparse PK sampling, Ctrough and Cmax
concentrations were taken for Cycles 1, 4 and 8. A further subgroup of Cohort 1 of 54 patients
(approximately 20 patients per treatment arm), underwent more extensive PK sampling during Cycle 4
and pharmacodynamic sampling during Cycle 1 (Cohort 2).
The analysis sets used for the analyses of data locked on 30-Sep-2015 are summarised in Table 13.
Table 13 - Analysis sets by treatment (Randomized set)
Assessment report
EMA/303207/2017 Page 43/115
Two PK analysis sets were defined for this analysis:
The PAS+A1 (N=239) consisted of all patients in Cohort 1 (including Cohort 2 patients), who
received at least one (partial or complete) dose of investigational treatment (Rixathon or
MabThera) and had at least one evaluable PK sample collected and analysed. The PAS+A1 was
used to present the descriptive statistics for Cmax and Ctrough of Rixathon and MabThera
during Cycles 1, 4 and 8. Observed values of Cmax (end of infusion, EOI) and Ctrough (pre-
dose) were reported. Particularly, Cmax and Ctrough at steady state (Cycle 4) were assessed
as secondary objectives.
The PAS+A2 (N=49) consisted of all patients in Cohort 2, who received the investigational
treatment (Rixathon or MabThera) and provided at least one evaluable PK sample after the
Cycle 4 dose. A non-compartmental PK analysis of serum concentration time profile of Rixathon
and MabThera was conducted based on PAS+A2 to obtain the AUC0-tau (AUC0-21days) and
AUCall during Cycle 4 (near steady state). This subgroup also underwent PD sampling during
Cycle 1.
Results
Baseline characteristics
Overall, baseline demographic and background characteristics were well balanced for the Rixathon and
MabThera treatment arms and reflected the intended target population for the study.
The demographics were similar between the treatment arms in terms of age, gender, race or body
surface area (BSA). The mean age of all patients was 56.9 years ranging between 23 to 84 years (57.5
years in Rixathon and 56.4 years in MabThera) and 46.1% of patients were ≥ 60 years of age (47.8%
in Rixathon and 44.4% MabThera). More than half (58% in Rixathon and 53.7% MabThera) of patients
in the FAS were female and approximately two-third (68.6% in Rixathon and 65.7% MabThera) of the
total patients were Caucasian. The majority of patients (57.4% in Rixathon and 55.6% MabThera) had
an Eastern Cooperative Oncology Group (ECOG) status of 0.
Secondary objective (GP13-301 Study), PAS + A1
The geometric mean and (CV% geometric mean) of the secondary endpoint Cmax at Cycle 4 Day 1
was 333.59 mcg/mL (41.09%) and 331.93 mcg/mL (35.32%) for the Rixathon and MabThera arms,
respectively (Table 14). In addition to Cmax in Cycle 4, Cmax in Cycles 1 and 8 were also evaluated.
The geometric means ratio of Cmax at Cycle 4 Day 1 (Rixathon /MabThera) was 1.00 and the
corresponding 90% CI was (0.925, 1.09). Based on the results, the Cmax at Cycle 4 Day 1 (near
steady state) was similar between the Rixathon and MabThera arms. Also Cmax at Cycle 1 Day 1 and
Cycle 8 Day 1 were similar between the Rixathon and MabThera arms.
Assessment report
EMA/303207/2017 Page 44/115
Table 14 - Summary of rituximab Cmax (mcg/mL), by treatment (Cohort 1) – study GP13-301 (PAS+A1)
The secondary endpoint, mean Ctrough levels at Cycle 4 Day 1 evaluated in Cohort 1 was 66.42
mcg/mL (CV% 71.7 %) and 82.13 mcg/mL (CV% 74.9 %) for the Rixathon and MabThera arms,
respectively. The Ctrough levels are similar in Cycle 4 Day 1 (around steady state) and Cycle 8 Day 1
and between Rixathon and MabThera and the data ranges in both arms largely overlap. The CV% was
high (greater than 70% in Cycle 4) and this was due to the fact that many patients have Ctrough
levels below the limit of detection and were reported as 0.
Assessment report
EMA/303207/2017 Page 45/115
Table 15 - Summary of rituximab Ctrough (mcg/mL), by treatment (Cohort 1) – study GP13-301 (PAS+A1)
GP13-301 Study: Secondary objective; PAS + A2
The geometric mean (CV% geometric mean) for AUC(0-21d) during Cycle 4 evaluated in Cohort 2 was
3210 mcg*day/mL (27.5%) and 3340 mcg*day/mL (34.9%) for the Rixathon and MabThera treatment
arms, respectively. Since many patients had rituximab concentrations below LLOQ at later time points
of the Cycle 4 PK profile (e.g., 360 h after EOI or pre-dose Cycle 5), their AUC(0-21d) could not be
accurately determined. In such cases, area under the curve from the time of dosing to the time of the
last observation, regardless of whether the last concentration was measureable or not (AUCall) was
used for exposure evaluation. The geometric mean (CV% geometric mean) for AUCall during Cycle 4
evaluated in Cohort 2 was 2510 mcg*day/mL (55.1%) and 2310 mcg*day/mL (109.1%) for the
Rixathon and MabThera treatment arms, respectively. Based on the results, the AUC(0-21d) and
AUCall are similar between the Rixathon and MabThera arms.
Assessment report
EMA/303207/2017 Page 46/115
Table 16 - Summary of AUC(0-21d) (mcg*day/mL) and AUCall (mcg*day/mL) for Cycle 4 (Cohort 2) - study GP13-301 (PAS+A2)
Figure 9 - Individual concentration-time profiles for rituximab by treatment (Cohort 2) (PAS+A2)
Special populations
Table 17 - PK data in patients > 65 years old
Age 65 – 74 (Older
subjects number / total
number)
Age 75 – 84 (Older
subjects number / total
number)
Age 85+ (Older subjects
number / total number)
Rixathon Originator Rixathon Originator Rixathon Originator
PK Trials
GP13-101 1/6 -- -- -- -- --
GP13-201 22/133 35/179 6/133 3/179 -- --
Assessment report
EMA/303207/2017 Page 47/115
2.4.3. Pharmacodynamics
Mechanism of action
No clinical pharmacodynamic studies on the mechanism of action have been submitted.
Primary and Secondary pharmacology
B-cell depletion as a marker for the effect of rituximab
Analytical methods
Peripheral B-cell counts were measured after immunostaining by flow cytometry assay. As CD19+ B-
cell depletion is considered a surrogate for CD20+ B-cell depletion, CD19+ analysis on B-cells has been
validated as an exploratory biomarker of Rixathon /MabThera activity by Quintiles and Eurofins.
PD results; Key secondary PD endpoint: B-cell depletion – AUEC(0-14d) (GP13-201 Study Part I)
In addition to evaluating PK bioequivalence, the GP13-201 study also evaluated equivalence in terms
of depletion of peripheral B-cells in response to Rixathon or MabThera defined as “area under the
effect-time curves” (AUECs) of the percent change of the blood B-cell count relative to baseline up to
the second infusion (i.e. Day 15). Other supportive PD endpoints were also evaluated as described
below.
In order to conclude equivalence, the 95% CI of ratio of the geometric means (Rixathon /MabThera)
had to be within the pre-specified equivalence limits of [0.8, 1.25].
The ratio of the geometric means (Rixathon /MabThera) was 1.019 and the corresponding 95% CI
[0.997, 1.042] was within the pre-specified equivalence limits. Therefore, the secondary endpoint for
equivalence was met.
Table 18 - AUEC(0-14d) of percent B-cells relative to baseline: Comparison between Rixathon and MabThera using ANOVA (GP13-201 Study Part I) (PAS)
Assessment report
EMA/303207/2017 Page 48/115
Additionally, the study evaluated the percentage of peripheral blood CD20+ B-cell levels relative to
baseline following Rixathon or MabThera infusion at Days 15, Week 12, 16, and 24 and proportion of
patients in whom the peripheral blood CD20+ B-cell counts decreased below the detection limit at
Days 1 (before first infusion), 4, 8 and 15 (before second infusion). The mean percentage of B-cell
counts relative to baseline over time was also similar between the two treatments.
Figure 10 - Arithmetic mean (SD) of percent B-cell relative to baseline over 52 weeks by treatment - (GP13-201 Study Part I) (PAS)
Overall, the proportion of patients with B-cell counts below the lower limit of quantification (LLOQ) was
comparable between Rixathon and MabThera on Days 4, 8 and 15. More than 50% patients by Day 8
and more than 70% patients by Day 1 5 were below the LLOQ in both treatment groups.
Table 19 - Summary of absolute peripheral blood B-cells below limit of quantification by treatment - (GP13-201 Study Part I) (PAS)
Peripheral CD20+ B-cells percentage relative to baseline (study GP13-301)
In addition to evaluating efficacy of Rixathon compared to MabThera in FL patients, the study
evaluated peripheral CD20+ B-cell depletion in terms of AUEC(0-21d) of percent change in blood
CD20+ B-cell count relative to baseline following the treatment with Rixathon -CVP and MabThera-CVP.
This study was not powered for PD analysis and all PD data were presented using descriptive statistics.
Pharmacodynamics (PD) analyses were conducted using Pharmacodynamic Analysis Set (PDAS).
Assessment report
EMA/303207/2017 Page 49/115
The PDAS consisted of 48 patients, 24 in the Rixathon and MabThera arms, respectively. The PDAS
included patients in Cohort 2 who received Cycle 1 of investigational drug and had at least one post-
baseline pharmacodynamic sample collected and analysed.
The Pharmacodynamic Analysis Set (PDAS) was used to analyze AUEC(0-21d) of all patients who
received the investigational treatment (Rixathon or MabThera) during Cycle 1 in Cohort 2. Blood
samples for this PD assessment were taken at screening and at the following time points during Cycle
1: pre-dose, EOI, 24 hours after EOI, 72 after EOI, and prior to Cycle 2. The baseline was defined as
the mean of peripheral CD20+ B-cell counts at screening and predose of Cycle 1. For each patient, the
percentage relative to baseline for peripheral CD20+ B-cell counts was calculated. The PD profile in
terms of peripheral CD20+ B-cells percentage relative to baseline was similar between Rixathon and
MabThera.
Table 20 - Summary of peripheral CD20+ B-cells percentage relative to baseline, by treatment - study GP13-301 (PDAS)
Assessment report
EMA/303207/2017 Page 50/115
Figure 11 - Arithmetic mean (SD) plot of percentage relative to baseline for peripheral
CD20+ B-cells counts vs times, by treatment (Cohort 2) - study GP13-301 (PDAS)
AUEC(0-21d) for peripheral CD20+ B-cell counts in Cycle 1 was a secondary endpoint, which was
calculated using the percentage relative to baseline data for peripheral CD20+ B-cell counts. The mean
AUEC(0-21d) during Cycle 1 evaluated in Cohort 2 was 1830 %*day (CV% 18.7%) and 1920 %*day
(CV% 12.0%) for the Rixathon and MabThera arms, respectively.The geometric means ratio of
AUEC(0-21d) (%*day) of peripheral B-cells (Cohort 2) (Rixathon /MabThera) was 0.939, and the
corresponding 90% CI was [0.845; 1.04]). The summary results show that the AUEC(0-21d) in Cycle 1
is similar between the two treatment arms.
Table 21 - Summary of AUEC(0-21d) (%*day) and Tlast (day) of peripheral CD20+ B cells
(Cohort 2) - study GP13-301 (PDAS)
Assessment report
EMA/303207/2017 Page 51/115
Immunogenicity
No special studies were conducted with immunogenicity assessment as primary objective under
Rixathon clinical development program; however, GP13-201 study in patients with RA, GP13-301 study
in patients with FL and GP13-101 study in Japanese patients with indolent NHL contributed to the
immunogenicity data for immunology and oncology set-up.
Bioanalytical methods for immunogenicity assessment:
The immunogenicity evaluation of rituximab as determined by the formation of antibodies against
rituximab was made by a 3-step procedure comprising a validated screening and confirmatory as well
as a titer assay based on an ELISA (used for RA study GP13-201) or bridging
electrochemiluminescence assay (ECL) format (used for FL studies GP13-101 and GP13-301) and a
validated neutralising antibody (NAb) assay. For the ADA screening, confirmatory and titer assays,
anti-rituximab antibodies were used as positive control. For the NAb assay, a neutralising anti-
rituximab antibody was used.
Immunogenicity in RA patients (GP13-201 Study Part I)
A total of 173 patients (86 in Rixathon and 87 in MabThera arms) were evaluated for immunogenicity.
According to the protocol, samples to assess the presence of anti-drug antibodies (ADA) were collected
at visits 3 (predose), 8 (14 days post dose), and 11 to 14 (96, 154, 252, 350 days post dose) and in
case of retreatment (pre dose) and at the 26 week retreatment follow up:
At randomization (pre-treatment), ADAs were detected in 2 (1.2%) patients in MabThera arm and
these 2 patients were excluded from further ADA summary. However, none of these 2 patients had
ADAs detected at later time points. Interestingly, ADAs to rituximab were also detected in 3.6% of
placebo-treated patients with active RA in the SERENE study (Emery et al 2010). Overall, post-
treatment ADA was detected in 9 out of 82 (11.0%) patients in the Rixathon and 18 out of 84 (21.4%)
patients in the MabThera arm.
Samples with confirmed positive ADAs were further assessed for the neutralizing capacity of the
antibodies using a cell based assay (NAb assay). A total of four patients exhibited NAbs, 3 out of 82
(3.7%) patients in the Rixathon arm and 1 out of 84 (1.2%) patient in the MabThera arm. Among
them, two patients, one in the Rixathon arm and one in the MabThera arm showed ADA titer > 100
mcg/mL. The highest ADA concentration (>2000 mcg/mL) in patients with NAb was detected for the
positive determined patient in the MabThera arm.
Immunogenicity in FL patients (study GP13-301)
In this study, immunogenicity (ADA) was assessed for all patients at screening (or pre-dose or both),
End of treatment (EOT) Combination Phase, and EOT Maintenance Phase. For Cohort 1 patients, an
additional immunogenicity assessment was performed at pre-dose Cycle 4 Day 1. Patients with pre-
existing immunogenicity were excluded from this assessment.
Overall, five patients with pre-existing ADAs at screening, were excluded from this assessment. Four of
these patients (three in Rixathon arm and one in MabThera arm) were consistently ADA negative at
later visits whereas one patient in the Rixathon group had confirmed non-neutralising ADA positive
result at the end of the maintenance phase.
Assessment report
EMA/303207/2017 Page 52/115
A total of 551 (87.6%) patients were included in the immunogenicity analysis, 268 (85.4%) patients in
the Rixathon and 283 (89.8%) patients in the MabThera arm. Overall, the number of patients with
detectable post-dose ADAs was low in the study (n/N=8/551 (1.5%)), with no significant differences
between Rixathon (n/N=5/268 (1.9%)) and the MabThera (n/N=3/283 (1.1%)) arms. This is
comparable with data provided for NHL patients in the MabThera SmPC: 1.1% of patients with DLBCL
developed human anti-chimeric antibodies. Based on the limited PK data in ADA positive patients,
there is no indication that ADA is having an impact on PK exposure.
Overall, NAbs were detected in 2 out of 268 (0.7%) patients in the Rixathon group and 2 out of 283
(0.7%) patients in the MabThera group. Among them, one in the Rixathon group was detected at the
end of Maintenance Treatment Phase. All other NAb positive incidences were detected during the
combination phase. An assessment of safety parameters found that all four patients had no AEs of
infusion-related reactions. One patient in MabThera group died due to cardiac arrest and had several
SAEs of which two SAEs (neutropenia) were considered to be related to study drug. This patient also
had grade 3 chills after the first infusion. Overall, all four incidences of NAb showed no link with any
immunogenicity related SAEs or diminished efficacy.
2.4.4. Discussion on clinical pharmacology
The analytical methods used are appropriate and well-described. In general, the pre-specified criteria
for the different validation parameters are in line with the Guideline on bioanalytical method validation
and were fulfilled.
From the pharmacokinetics point of view, the PK pivotal Study GP13-201-Part 1 and supportive study
GP13-301 are considered sufficient to demonstrate biosimilarity between the test and reference
product. Supportive Study GP13-101 is less relevant as the rituximab product used is Rituxan licensed
in Japan- considering no bridge between the EU Mabthera and Rituxan (Japan) has been established.
The number and type of studies submitted are in line with EMA Guideline on similar biological medicinal
products containing biotechnology-derived proteins as active substance: non-clinical and clinical issues
(EMEA/CHMP/BMWP/42832/2005 Rev 1).
The biosimilar comparability limit were met for the primary PK parameter AUC(0-inf) from RA patients
in study GP13-201 Part I. Uncertainties about the exclusion of ADA positive patients from PK
evaluation and the calculation of adjusted mean values were discussed, however there was no relevant
difference observed in the PK exposure of ADA positive RA patients compared to patients who did not
develop ADAs in study GP13-201. Considering that the (geometric) mean and median of the half-life of
the two products are very similar, a statistically significant difference in AUC values by ADA status was
unlikely.
In study GP13-201 Cmax1 slightly exceeded the upper limit (i.e. 126%), this was due to the high
variability caused by longer duration and more escalation steps during the 1st infusion in comparison
to the 2nd infusion. It is agreed that for a drug product delivered via infusion, the peak systemic
concentrations represent the end-of-infusion concentrations which are highly dependent on infusion
rate and are therefore not adequate as a measure for bioavailability. Cmax2 was within the standard
bioequivalence limits [0.8, 1.25]. Therefore, the criterion of bioequivalence was met.
Due to the sparse sampling of PK data in study GP13-301 in FL patients, the value of these data for
the decision on PK similarity is limited and only supportive. The descriptive PK results do not fully
support PK similarity of Ctrough at predose cycle 4 (mean Ctrough 66.4 vs. 82.1 mcg/mL for Rixathon
and MabThera, resp.) with Rixathon exposure being around 10% lower than in the MabThera group.
On the other hand, comparison of pre-dose cycle 8 Ctrough values (123 vs. 127 µg/mL for Rixathon
Assessment report
EMA/303207/2017 Page 53/115
and MabThera, respectively) supports similarity between the groups. Frequency of BLQ samples in both
treatment arms was low in both ADA positive and ADA negative patients and comparable between the
treatment groups. This indicated that Ctrough results from cohort 1 are not biased by BLQ treatment.
The comparison of AUC(0-21d) in Cohort2 (subgroup with full profile) was also influenced by the fact
that many patients (around 25%) had rituximab concentrations below LLOQ at later time points of the
Cycle 4 PK profile (e.g., 360 h after EOI or pre-dose Cycle 5) so that their AUC(0-21d) could not be
accurately determined. However, AUC comparison in the remaining patients (n=20 and 17) in cohort 2
rather supports similarity and does not indicate a large difference in bioavailability of Rixathon and
MabThera during steady state in the presence of co-medication. Since the difference between AUCall
(area under the curve from the time of dosing to the time of the last observation, regardless of
whether the last concentration was measureable or not) and AUC(0-21d) was the same in both groups,
observation and handling of BLQ samples appeared to have no relevant impact on the supportive
evidence for biosimilarity.
PK data from the rheumatoid arthritis and follicular lymphoma patients are being extrapolated to the
other oncology indications (i.e. chronic lymphocytic leukaemia) and auto-immune indications (i.e.
granulomatosis with polyangiitis and microscopic polyangiitis). In all indications, the clearance of
rituximab occurs through two pathways: the non-specific IgG clearance pathways and the target
mediated drug disposition (TMDD) pathway. The non-specific IgG clearance pathways are reported to
be mediated by binding to the neonatal Fc receptor (FcRn) and to various Fc receptors (FcγRs).
Rixathon and MabThera, being similar in structure, bind to FcRn and various FcγRs with similar affinity.
The TMDD pathway for rituximab is mediated by binding to the CD20 receptor on target cells. It is
known that the extent of TMDD in various populations is correlated to the CD20+ B-cell levels (or
tumour load in haematologic oncology indications), which differ within and between populations (e.g.
CD20+ B-cell NHLs << CLL). Despite the expected difference in CD20+ B-cell levels and extent of
TMDD in different disease indications, the same clearance mechanisms (non-specific and TMDD) is
expected. In addition, the comparable in vitro binding properties of Rixathon combined with the human
comparability data in B-cell depletion from both the autoimmune (RA) and oncology (FL) indications,
provide sufficient evidence that the clearance of Rixathon is expected to be comparable to that of
MabThera in all indications that are approved for MabThera. Therefore, an extrapolation of the PK data
from rheumatoid arthritis and follicular lymphoma patients to the other oncology and auto-immune
indications of MabThera can be considered acceptable.
The discriminative power of the PD parameter AUEC(0-14d) of B-cell depletion (based on 3 data
points) is considered low, since B-cell depletion was almost complete from day 3 to week 12 in both
arms of study GP13-201. The similar time courses of B-cell recovery after 16 weeks can be considered
as supportive for similarity, since the time interval of B-cell recovery is considered to be quite sensitive
for detecting differences. On the other hand, potential differences might be more indicative for other
(patient) factors than for PK differences. In contrast to the PK analysis, ADA positive patients were
included in PD evaluation of study GP13-201. The individual time courses of B cell depletion by ADA
status showed that inclusion of ADA positive patients did not have a significant impact on the
evaluation of the time course of B cell recovery. In study GP13-301, B-cell depletion was complete in
all samples over the whole treatment cycle, thus, the discriminative power of the parameter AUEC(0-
21d) for detecting differences is also low.
As far as known, there is no strong correlation between the extent of B-cell reduction and the extent of
the clinical response in RA. For NHL, the correlation is even less clear, since circulating B-cells may not
directly reflect tumour mass, and this response cannot be considered as an appropriate surrogate of
the clinical response. Nevertheless, the comparative B-cell depletion data are considered relevant for
Assessment report
EMA/303207/2017 Page 54/115
the assessment of bio-similarity and extrapolation to other non-investigated indications, since the B-
cell levels indirectly reflect the potency of the drug of antibody dependent cellular cytotoxicity (ADCC),
complement dependent cytotoxicity (CDC), and programmed cell death; these mechanisms of action
are applicable to all indications.
As part of immunogenicity assessment the following aspects were considered, according to EMA’s
Guideline on Immunogenicity assessment of biotechnology-derived therapeutic proteins: i) rate of
ADAs; ii) hyper-acute / acute reactions; iii) delayed reactions and iv) autoimmunity (cross-reactivity to
an endogenous counterpart) (See Clinical safety).
Two different assay formats were used for the determination of immunogenicity (anti-drug antibodies,
ADA) in the sera of the two pivotal studies (affinity capture elution ELISA method for the RA patient
serum samples and the ECL assay for the FL serum measurements) and 2 approaches for the detection
of neutralizing anti-rituximab antibodies (NAb assay) were applied: In RA study GP13-201, a cell-based
CDC assay format was chosen, whereas in the FL study, due to very strong matrix interference in FL
serum, the assay format had to be changed to a competitive ligand binding assay format.
Both immunogenicity screening assays are considered appropriate to detect ADAs, however, the assay
in FL serum is considered more reliable than the RA assay in terms of sensitivity and drug tolerance.
Both assay formats are designed to measure Rixathon -specific ADAs rather than MabThera-specific
ADAs. No data were provided on the use of the originator substance rituximab as capture and bridging
antigen which could demonstrate antigenic equivalence of Rixathon and MabThera for binding of ADAs
in the assays. The chosen design may overestimate ADA incidence of Rixathon vs. MabThera rather
than vice versa, it therefore does not give a competitive edge for the biosimilar product. From a
comparison of drug concentration present in the immunogenicity samples, it can be concluded that the
majority (75%) of NAb incidence results can be considered as conclusive.
There are only few data on the incidence of ADA after intravenous application of MabThera in RA
patients from other clinical studies for comparison: Similar (slightly lower) incidences of treatment-
emergent ADA (13,5 % and 17.6 %) were observed in recent biosimilar PK trials for the RA patient
group receiving MabThera (Cohen et al. 2016 and Yoo et al. 2013). The longer observation period in
the Rixathon study (including the time period of B cell recovery) compared to the literature data is
considered as supportive for a proper immunogenicity assessment and might be a reason for
observation of higher incidences. It is agreed that comparison between studies in literature may be
due to different assays used. At any rate, it is likely that the difference of -10% could be a chance
finding and ultimately it is not of concern since immunogenicity –if anything—is lower than that of the
reference product which is in accordance with the overarching guidelines (CHMP/437/04 Rev 1;
EMEA/CHMP/BMWP/42832/2005 Rev1).
Immunogenicity results revealed the ADA incidence being 2-fold lower in RA patients treated with
Rixathon compared to MabThera (11.0 % vs. 21.4 %). The statistical significance and potential
reasons of this finding was questioned. The incidences lie in the range of treatment-emergent ADA
incidences (13.5 % and 17.6 %) observed in recent trials in RA patients receiving MabThera (Cohen et
al. 2016 and Yoo et al. 2013). In the oncology setting a lower immunogenicity of Rixathon (2.1%) and
MabThera (0.9%) was observed; however, significance of this result is limited. The observed number
of patients with ADA (including NAb) in trial GP13-301 was very low which can probably attributed to
the immunosuppressive effect of the concomitant chemotherapy. Therefore, this population is not the
most sensitive population for detecting a difference in immunogenicity. ADA incidence for MabThera in
this study was similar to the 1.1 % observed diffuse large B cell lymphoma (DLBCL) patients receiving
MabThera 375 mg/m2 plus standard CHOP chemotherapy every 3 weeks for eight cycles (see
MabThera SmPC). In the SAWYER trial (BO25341) in CLL patients receiving MabThera in combination
Assessment report
EMA/303207/2017 Page 55/115
with fludarabine and cyclophosphamide (FC), the incidence of treatment-induced anti-rituximab
antibodies against MabThera was slightly higher (6.7%) in the intravenous arm and 2.4% in the
subcutaneous arm (see Mabthera SmPC).
2.4.5. Conclusions on clinical pharmacology
Comparison of PK parameters and the time courses of B-cell depletion as PD biomarker in both studies
shows biosimilarity of Rixathon to Mabthera, which is supported by the similar time courses of B-cell
recovery after 16 weeks (and by ADA status).
An extrapolation of the PK/PD data from rheumatoid arthritis and follicular lymphoma patients to the
other oncology and auto-immune indications of MabThera is acceptable due to the same clearance
mechanisms (non-specific and target-mediated), similar in vitro binding properties of Rixathon and
comparable human data in B-cell depletion from both the autoimmune (RA) and oncology (NHL)
indications.
2.5. Clinical efficacy
2.5.1. Dose response studies
Dose response studies were not submitted (see discussion on clinical efficacy).
2.5.2. Main study
Study GP13-301 - Pivotal clinical efficacy and safety study in patients with FL
Methods
Study GP13-301, conducted in patients with previously untreated, advanced stage FL is a randomised,
active-controlled, double-blind, multi-center, parallel group confirmatory study to compare the efficacy,
safety, PK, and pharmacodynamics of Rixathon plus CVP vs MabThera plus CVP in patients with
previously untreated, advanced stage FL.
Patients who achieved a CR or PR at the end of the combination treatment period were eligible to
receive single-agent Rixathon /MabThera maintenance treatment, according to original treatment
assignment, for a period of two years. Patients for whom progression of disease was documented at
any time during either the combination or maintenance treatment periods, discontinued study
treatment, but continued to be followed for survival every 6 months for a total period of a maximum of
three years from the date of randomisation.
Assessment report
EMA/303207/2017 Page 56/115
Note: during maintenance period, in Italy only, Rixathon /MabThera every 2 months for 2 years (12 cycles) or until disease progression AE=Adverse event; CVP= Cyclophosphamide, vincristine, prednisone; FLIPI=Follicular Lymphoma
International Prognostic Index; PFS=Progression free survival; q=every “x” months or years Methods Figure 12 - Overall GP13-301 study design
Study participants
The study population consisted of adult (≥ 18 years old) patients with previously untreated, advanced
stage (Ann Arbor stage III/IV) FL of WHO histological grades 1 -3a. FL histology, WHO histological
grade, and CD20-positivity were confirmed by central pathological testing prior to patient
randomization.
Man Inclusion criteria
Patients eligible for inclusion in this study had to meet all of the following criteria:
1. Patient with previously untreated advanced stage, CD20-positive FL:
a. Ann Arbor classification stage III/IV; b. WHO histologic grade 1, 2 or 3a, as confirmed by central pathological testing; and c. Require therapy for FL as per local guidelines or in the opinion of the treating
physician. 2. Patient age ≥ 18 years. For India only as per Protocol Amendment 5: patients aged 18 to 75
years, inclusive 3. Patient with at least one measurable lesion (accurately measureable in at least 2 perpendicular
dimensions); a. at least 1 measurable nodal lesion > 20 mm in the long axis; OR
b. at least 1 measurable extranodal lesion with both long and short axes ≥ 10 mm.
Assessment report
EMA/303207/2017 Page 57/115
Main Exclusion criteria
Patients eligible for this study must not have met any of the following criteria:
1. Patient with grade 3b (aggressive) FL or any histology other than FL grade 1, 2, or 3a. 2. Patient with histological evidence of transformation to high grade or diffuse large B-cell
lymphoma.
3. Patient who has previously received any prior therapy for lymphoma, e.g., cytostatic or cytotoxic agents, antibodies, anti-lymphoma vaccination, experimental treatments and radiotherapy, except who received involved field radiation 4 weeks prior to Cycle 1 Day 1, of up to two lesions that will not be used to evaluate disease progression.
4. Evidence of significant leukemic disease, defined as >10 x 109 /L circulating CD20+ lymphoma cells.
5. Patient with clinical evidence of central nervous system involvement by lymphoma or any
evidence of spinal cord compression by lymphoma. 6. Patient with evidence of any uncontrolled, active infection (viral, bacterial, including
tuberculosis or fungal). 7. Patient receiving chronic (>3 months high dose (>20 mg of prednisone or > pproximately 3
mg of dexamethasone per day or equivalent doses of other steroid medications) of systemic
corticosteroids.
8. Patient with any malignancy within 5 years prior to date of randomization, with the exception of adequately treated in situ carcinoma of the cervix uteri, basal or squamous cell carcinoma, or non-melanomatous skin cancer.
9. Patient with a known hypersensitivity to any of the study treatment ingredients, e.g., to recombinant human antibodies.
10. Patient with concurrent serious illnesses, uncontrolled medical conditions, or other medical history including clinically relevant abnormal laboratory results, which in the investigator’s
opinion would interfere with a patient’s participation in the study, or with the interpretation of study results:
a. uncontrolled neurological disease (e.g., recurrent seizures despite existing anticonvulsant therapy);
b. neuropathy ≥ grade 1, neuromuscular disease; c. severe disturbance of liver function; d. severe constipation;
e. cystitis or other ongoing infections; f. disturbance of micturition; g. severe chronic obstructive pulmonary disease with clinically manifest hypoxemia;
h. uncontrolled hypertension (defined as systolic blood pressure (BP) > 160 mm Hg, or diastolic BP > 100 mm Hg);
i. history of stroke or cerebral ischemia (within 6 months prior to screening );
j. history of myocardial infarction or other clinically significant myocardial disease (within 6 months prior to screening or unstable angina (≥ New York Heart Association Grade II);
k. known infection with human immunodeficiency virus (HIV) or any other severe immune-compromised state according to patient history (if required by local regulations or clinical practice guidelines, patient may be tested during the screening period to confirm HIV status);
l. evidence of ongoing drug or alcohol abuse within the last 6 months before screening; 11. Patient has had major surgery, open biopsy or trauma within 4 weeks prior to date of
screening (lymph node biopsy is not regarded as major surgery), or expects the need for major surgery during the course of study treatment.
12. Female patient who is nursing (lactating/breast-feeding), pregnant or planning a pregnancy within 12 months after the last infusion of study drug; where pregnancy is defined as the state of a female after conception and until the termination of gestation, confirmed by a positive hCG
laboratory test (> 5 mIU/mL). 13. Patient has received therapy with any other investigational medicinal product within the last 30
days or 5 times the half-life, whichever is longer, prior to screening. 14. Patient plans to receive live vaccines during the study or has received live vaccines 4 weeks
prior to date of screening. Examples of live vaccines include intranasal influenza, measles, mumps, rubella, oral polio, BCG, yellow fever, varicella, and TY21a typhoid vaccines.
15. Patient is using growth factors or transfusions to meet study eligibility requirements during Screening period. (The use of growth factors and transfusions during screening is permissible,
Assessment report
EMA/303207/2017 Page 58/115
if there is suspicion of bone marrow involvement by lymphoma and patient is deemed not to be growth factor-dependent or transfusion-dependent).
Treatments
Study treatment (drug) includes any of the drugs as follows: Rixathon, MabThera, CVP, or combination
of these drugs in either of the two treatment arms administered to the patient as part of the required
study procedures.
Table 22 - Combination Treatment Period – Dosing Schedule by Treatment
Figure 13 - Study design – combination and maintenance periods
Infusion procedures for Rixathon and MabThera
The following recommended pre-medications may have been given 30 minutes prior to the start of
infusion for all combination treatment cycles paracetamol (500 mg p.o.); H1 antihistamine (p.o. or
i.v.); prednisone (100 mg p.o.).
Assessment report
EMA/303207/2017 Page 59/115
Objectives
Primary objective:
To demonstrate comparability of the overall response rate (ORR) in patients with previously untreated,
advanced stage FL who receive Rixathon -CVP combination treatment to patients who receive
MabThera-CVP combination treatment.
Other Efficacy objectives:
To evaluate complete response (CR) rate
To evaluate partial response (PR) rate
To evaluate progression-free survival (PFS)
To evaluate overall survival (OS)
Safety objectives:
To describe safety of Rixathon in comparison to MabThera either as a single agent or in
combination with CVP
To evaluate the incidence of immunogenicity (anti-drug antibody [ADA] formation) against
Rixathon and MabThera
Pharmacokinetic (PK)/Pharmacodynamic objectives:
To evaluate the PK of Rixathon and MabThera
To evaluate a Pharmacodynamic marker following the treatment with Rixathon -CVP and
MabThera-CVP
Outcomes/endpoints
Efficacy endpoints included ORR, CR, PR, PFS, and OS. Efficacy was evaluated using Modified Response
Criteria for Malignant Lymphoma based on contrast-enhanced Computerized tomography (CT) or
Magnetic Resonance Imaging (MRI) and bone marrow evaluations. B-symptoms were also evaluated at
the same time points as tumour assessments.
Sample size
Anticipating an expected overall response rate of 81% for both treatment arms and a pre-specified
equivalence margin of 12%, it was calculated that a total of 556 patients (288 per arm) were required
to assess equivalence of Rixathon to the reference drug at a two one-sided significance level of 2.5%
with 90% power. In order to allow for 10% of drop-outs and major protocol violations, a total of 618
patients were planned to be randomized; 629 patients were actually randomized into the study.
The equivalence margin (δ) was based on historical data of a phase III study in patients with
previously untreated advanced FL (Ann Arbor classification stages III to IV), where the patients
received either CVP plus MabThera (n=162) or CVP (n=159) (Marcus 2005). The observed ORR in the
study was 81% for the MabThera-CVP arm and 57% for CVP arm, the observed add-on effect of CVP
plus MabThera on ORR was 24% (95%-CI: 14% to 34%). The equivalence margin of 12% was
determined considering the variability of the point estimate of the add-on effect (24%) by taking a
value lower than the lower bound of the 95% CI for MabThera-CVP vs. CVP obtained from the historical
data.
An approximate cohort of 100 patients per treatment arm (Cohort 1, who consented to the trial)
underwent PK sampling. A further subgroup of approximately 20 patients per arm (Cohort 2, i.e., 40 of
the patients in Cohort 1), also were to undergo extensive PK sampling during Cycle 4 and
pharmacodynamic sampling during Cycle 1. Immunogenicity (ADA) formation) samples were collected
from all patients.
Assessment report
EMA/303207/2017 Page 60/115
Figure 14 - Planned PK/Pharmacodynamic cohort design
Randomisation
Subjects were randomised (via IRT) in an 1:1 ratio to receive either Rixathon -CVP or MabThera-CVP in
a 1:1. Randomisation was stratified by FLIPI risk group (FLIPI score 0-2 vs. FLIPI score 3-5),
geographic region (Asia Pacific, Latin America, and Europe) and by participating cohort for PK data
collection (Cohort 2: extensive PK sampling; Cohort 1 exclusive of Cohort 2: sparse PK sampling; rest
of patients without PK sampling).
Blinding (masking)
Patients, investigator site staff, persons performing the assessments, and data analysts in this study
remained blinded to the identity of the treatment from the time of randomisation until database lock
(DBL). As the appearance of the test and reference product was not identical, the preparation of the
study medication was performed by dedicated unblinded site staff not involved in the further conduct
of the study. Unblinding was to occur only in the case of patient emergencies, for regulatory reporting
purposes, and at the conclusion of the study. Unblinding occurred when the primary DBL was reached
(i.e. when all randomized patients have completed the combination treatment period).
Statistical methods
In general data were summarised by statistical characteristics (categorical data: absolute and relative
frequencies; continuous data: mean, standard deviation, median, minimum, and maximum) stratified
by treatment group and visit (where applicable).
Efficacy analyses
The primary efficacy endpoint was ORR during the combination phase as based on Central Blinded
Review of radiological response. The primary analysis of ORR was based on the per-protocol set (PPS).
Equivalence was to be concluded in case the 95% CI for the difference of proportions in ORR between
both treatment groups fell completely into the pre-defined equivalence range (-12%, 12%). Normal
approximation to the binomial distribution was used to generate the 95% CI. In addition a logistic
regression model with the exploratory variables FLIPI score and treatment was used to derive an
Assessment report
EMA/303207/2017 Page 61/115
estimate and associated 90% CI of the odds ratio of treatment difference (Rixathon versus MabThera)
adjusting for FLIPI score.
To support the primary analysis the following analyses were planned:
analysis of CRB-ORR on the PPS excluding patients whose best overall response was unknown
or missing
CRB-ORR performed on FAS
sub-group analyses of ORR on both PPS and FAS by age and FLIPI score risk group
ORR based on investigator review for the PPS and FAS
For the primary endpoint, subjects with missing endpoint information were considered non-responder.
Time-to-event analyses were based on the Kaplan-Meier approach. For OS and PFS the median time
(including its 90%-CI) was estimated for each treatment arm. In addition, the 25% and 75%
percentiles were also provided for each treatment arm. Using the Kaplan-Meier methodology, the PFS
and survival probabilities at 6, 12, 24, 30, and 36 months including their 90%-CIs were estimated. The
hazard ratio between treatment arms was estimated from a Cox regression model with treatment as
covariate and FLIPI score as stratification factor. The associated 90% CI was also generated.
Safety analyses
Safety analyses were mainly based on the frequency of AEs and on the number of patients with
laboratory values that fell outside the pre-determined ranges. Other safety data (e.g. vital signs, ECG)
were summarized similarly.
PK analyses
Descriptive statistics stratified by treatment were presented for Cmax and Ctrough (pre-dose) during
Cycles 1, 4, and 8 for Cohort 1patients. Furthermore, for Cycle 4 Day 1 Cmax, two-sided 90% CIs for
the mean difference of Rixathon (test) versus MabThera (reference) on the logscale were calculated
and the back-transformed point estimate as well as the 90% CI for the ratio of geometric means were
also provided (for descriptive purposes only).
PD-analyses
Summary statistics stratified for treatment and visit were presented for B-cell counts, absolute
changes and percentage relative to baseline in CD19+ B-cell values. The depletion of peripheral B-cells
was measured as the area under the effect-time curve (AUEC(0-21d)) of the percent peripheral B cell
count relative to baseline up to the second infusion (i.e., Day 21 or Cycle 2 Day1). For AUEC0-21d,
two-sided 90% CIs for the mean difference of Rixathon versus MabThera (test - reference) on the log-
scale was calculated and the backtransformed point estimate as well as the 90% CI for the ratio of
geometric means was provided (for descriptive purposes only).
Results
Participant flow
The planned number of randomised patients for this analysis was 618.
Assessment report
EMA/303207/2017 Page 62/115
However, 629 were actually randomised (314 patients to the Rixathon arm and 315 patients to the
MabThera arm; two Rixathon patients were mis-randomised and were discontinued before being
treated).
Table 23 - Patients’ disposition in the Combination Treatment Phase (Study GP13-301)
As of 10-Jul-2015 data cut-off date, 73.6% (n=231) of patients in the Rixathon arm and 73.3%
(n=231) of patients in the MabThera arm entered the Maintenance Phase and received at least one
dose of the maintenance treatment. A total of 303 patients were still receiving study treatment in the
Maintenance Phase, 142 patients (61.5%) in the Rixathon arm and 161 patients (69.7%) in the
MabThera arm. In the Maintenance Phase of the study, the primary reason for EOT (38.5% in Rixathon
arm and 30.3% in MabThera arm) was disease progression (16.0% Rixathon; 10.8% MabThera) and
“Treatment duration completed as per protocol” (15.2% Rixathon; 13.4% MabThera).
Recruitment
Study initiation date: 01-Dec-2011 (first patient screened)
Study completion date: 09-Jul-2015 (last patient last visit in the combination treatment period)
Study center(s): 174 centers screened patients and 159 centers randomized patients from centers in
26 countries.
Follow-Up period: 2.5 years (max. 3 years)
Conduct of the study
The study protocol was amended 5 times, the original protocol and all amendments are provided:
Amendment 1 (27-Feb-2012): the key purpose of this protocol amendment was to update and clearly
define the attributes of the study inclusion and exclusion criteria, revise the recommended dose
modification per current clinical practice, and change the primary ORR analysis from local to Central
Blinded Review of radiological response.
Assessment report
EMA/303207/2017 Page 63/115
Amendment 2 (17-Jan-2013) was a country specific amendment for Italy to change the schedule of
treatment in the maintenance phase of the study from every 3 months to every 2 months as requested
by Italian health authority.
Amendment 3 (21-Jan-2013) with the primary purpose to allow lymph node biopsy samples collected
within 5 months of screening to establish and confirm the diagnosis of follicular lymphoma to confirm
patient eligibility. Furthermore, the population PK of Rixathon and MabThera was removed in the
amendment as this analysis was not deemed necessary to compare the PK of Rixathon and MabThera.
Amendment 4 (26-Sep-2013), was a global amendment with the primary reason to add a planned
interim analysis (based on specific health authority feedback) to support the regulatory filing of
Rixathon with EMA and potentially other health authorities.
Amendment 5 (10-Nov-2014). This amendment was based on Health Authority feedback received in
February 2014, which lead to the removal of the planned interim analysis added in protocol
amendment version 4.0. As of 20-Oct-2014, 561 patients were enrolled in the study.
Baseline data
Table 24 - Demographics and baseline characteristics - study GP13-301 (FAS)
Assessment report
EMA/303207/2017 Page 64/115
Table 25 - Disease history and baseline characteristics by treatment (FAS)
Numbers analysed
The study was conducted in 26 countries and 159 centers randomised patients.
The planned number of randomised patients for this analysis was 618, however, 629 were actually
randomised (314 patients to the Rixathon arm and 315 patients to the MabThera arm; two Rixathon
patients were mis-randomised and were discontinued before being treated).
Assessment report
EMA/303207/2017 Page 65/115
Table 26 - Patient disposition by treatment – Combination phase (Randomized set)
Data sets analyzed
Table 27 - Analysis sets by treatment (Randomized set)
A total of 279 patients had at least one protocol deviation and these were balanced between the
treatment arms: 137 patients (43.9%) in the Rixathon arm and 142 patients (45.1%) in the MabThera
arm. These deviations were not believed to have introduced a bias in the efficacy or safety data
comparisons between the two treatment arms.
Outcomes and estimation
Primary efficacy results
The ORR, based on the Modified Response Criteria for Malignant Lymphoma using central blinded
review of the radiological response and liver/spleen enlargement assessments from the patients in the
Assessment report
EMA/303207/2017 Page 66/115
PPS, was 87.1% in the Rixathon arm and 87.5% in the MabThera arm and the difference in ORRs was
-0.40% (95% CI [-5.94%, 5.14%]; 90% CI [-5.10%, 4.30%]). Equivalence was concluded as the
entire 95% CI for the difference in ORR between the two treatments was within the pre-specified
equivalence margin of ±12%.
Of the 624 patients in the PPS 35 patients (5.6%; 19 Rixathon patients, 16 MabThera patients) had
missing/unknown best overall response (BOR) based on central blinded review. When these patients
were excluded from analysis, Overall response rate (CR or PR) was similar 0.55 (-4.03, 5.14) thus
confirming previous findings. Also, OR for FAS was similar with a difference of -0.44, (95% CI: 4.03,
5.14).
Table 28 - Primary efficacy analysis of overall response rate based on central blinded review (PPS)
Table 29 - Overall response rate based on central blinded review of tumour assessment by treatment (FAS)
Subgroup analyses of ORR were performed as supportive analyses to the primary analysis:
Subgroup analyses of ORR by FLIPI score and age (< 60 years vs. ≥ 60 years; PPS, FAS). Analyses
according to age demonstrated similarity. However, the two arms were numerically different for ORR
randomized by FLIPI strata, favoring MabThera (91.2% vs. 82.8% in Rixathon) in the subset of
patients with a FLIPI score 0-2 (low-intermediate risk), and favoring Rixathon (90.4% vs. 84.7%) in
the subset of patients with a FLIPI score 3-5 (high risk).
Secondary efficacy results
Best overall response (BOR)
In general, the proportions of patients with BORs for the four categories were similar for the two
treatment arms. The proportion of patients with BOR as ‘unknown’ was summarised by reason for the
combination phase in the PPS and the FAS. In general, the reasons for an unknown response were
balanced between the two treatment groups.
Assessment report
EMA/303207/2017 Page 67/115
Table 30 - Best overall response based on central blinded review (PPS)
Table 31 - Best overall response based on investigator versus central blinded review (PPS)
This table displays the number of patients who had a given BOR as per investigator (rows) and as per
central (columns), regardless of treatment arm.
PFS
PFS was based on investigator assessment only.
Assessment report
EMA/303207/2017 Page 68/115
The median PFS was not reached for either of the two treatment groups. As of the time of data cutoff
(10-Jul-2015), respectively 21.5% [n=67] and 16.5% [n=52] of patients in Rixathon and MabThera
treatment arms progressed or died, with 62.5% [n=195] and 71.7% [n=226] patients censored
without event and having adequate follow-up, respectively. The hazard ratio (HR) estimate and its
associated 90% CI are obtained by fitting Cox regression model with treatment allocation as covariate
and FLIPI score as stratification factor. The PFS HR (Rixathon /MabThera) was 1.33 (90% CI: [0.98,
1.80]), based on the investigator assessment.
Figure 15 - Kaplan-Meier plot of PFS by treatment based on investigator assessment (FAS)
Ancillary analyses
The PFS analysis was repeated with refined criteria for PFS events and censoring.
PFS – Set 1 is defined using the following modified censoring rules: If patient has no event, the date
is set as date of last adequate tumour assessment; If there is no adequate assessment, then the date
is set as randomisation date; An adequate assessment is defined as an assessment with known overall
response (complete response [CR], partial response [PR], stable disease [SD] or progressive disease
[PD]). As of database cut-off, 10-Jul-2016, 91 of 312 (29.2%) patients in the Rixathon and 78 of 315
(24.8%) patients in the MabThera treatment arms had a PFS event using the set 1 definition. The
result of the sensitivity analysis of PFS following the set 1 rules with a hazard ratio of 1.22 (with 90%
CI: [0.95, 1.58]).
Assessment report
EMA/303207/2017 Page 69/115
Table 32 - Sensitivity analysis of PFS based on investigator assessment: KM and cox-regression method with modified censoring rules – Set 1 (FAS)
Figure 16 - K-M plot of PFS by treatment (by investigator – modified censoring rules – Set 1
(FAS)
PFS – Set 2
PFS for set 2 is defined as the time from the date of randomisation to date of event which is defined as
the first observation of disease progression (documented or without documentation of evidence of
disease progression based on the investigator response), death, or start of another cancer therapy,
using the following modified censoring rules: If patient has no event, the date is set as date of last
adequate tumour assessment; If there is no adequate assessment (with known overall response CR,
PR, SD or PD), then the date is set as randomisation date.
Table 33 - analysis of PFS based on investigator assessment: KM and cox-regression – Set 2 (FAS)
Assessment report
EMA/303207/2017 Page 70/115
Figure 17 - K-M plot of PFS by treatment (by investigator – Set 2 (FAS)
Figure 18 - Kaplan-Meier plot of PFS by treatment based on investigator assessment (FAS) –
data cut off December 2016
Assessment report
EMA/303207/2017 Page 71/115
GAP analysis for PFS
In order to assess the robustness of the random censoring mechanism for PFS, for censored patients,
the gap between data cutoff date and last tumour assessment was assessed.
The median gap time for PFS follow-up as compared to cut-off date in patients who were still followed-
up for efficacy between the Rixathon and MabThera groups was 3.2 months and 3.0 months,
respectively. A minority of patients had a gap of more than one year between the data cutoff date and
their last tumour assessment; 8.2% [n=20] and 5.7% [n=15] of censored patients in Rixathon and
MabThera groups, respectively.
For all patients, 30.3% and 25.4% of patients had less than 6 months and 6 to 12 months PFS follow-
up, respectively; these percentages were similar between the treatment groups. Considering that the
majority of PFS events occurred during the Maintenance Phase, these figures are suggestive of
immaturity of PFS data and should be taken into account when interpreting PFS results.
OS
At the time of data cutoff (10-July 2015), the number of deaths (18 events in the Rixathon group vs.
17 events in the MabThera group) were similar between the two treatment arms.
Table 34 - Analysis of OS using Kaplan-Meier and Cox regression (FAS)
The median OS has not yet been reached for either treatment group based on the currently available
data. The analysis of OS using the Cox regression method showed that the HR was 1.03 (90% CI:
0.59, 1.80).
Assessment report
EMA/303207/2017 Page 72/115
Figure 19 - Kaplan-Meier plot of OS by treatment (FAS)
The analysis of overall survival using the Kaplan-Meier and Cox-regression method for the PPS was
consistent with the findings for the FAS.
The reasons for censoring patients were also similar between the two treatment groups.
Figure 20 - Kaplan-Meier plot of OS by treatment (FAS)- data cut off 31-12- 2016
Assessment report
EMA/303207/2017 Page 73/115
Table 35 - Summary of reasons for censoring patients for PFS and overall survival analysis
by treatment (FAS – data cut off July 2015)
Summary of main study
The following tables summarise the efficacy results from the main studies supporting the present
application. These summaries should be read in conjunction with the discussion on clinical efficacy as
well as the benefit risk assessment (see later sections).
Table 36 - Summary of efficacy for trial GP13-301
Title:
A randomised, controlled, double-blind Phase III trial to compare the efficacy, safety and
pharmacokinetics of Rixathon plus cyclophosphamide, vincristine, prednisone vs. MabThera plus
cyclophosphamide, vincristine, prednisone, followed by Rixathon or MabThera maintenance therapy
in patients with previously untreated, advanced stage follicular lymphoma
Study identifier GP13-301 (CIGG013A2301J)
Design Parallel group design
Duration of main phase: 8 cycles (~ 6 months)
Duration of Run-in phase: not applicable
Duration of Extension phase: not applicable
Hypothesis Equivalence
Treatments groups
Rixathon
Combination Phase: Rixathon: 375 mg/m2 + CVP every 21 days for 8 cycles; number randomized = 314 (2 misrandomized)
Assessment report
EMA/303207/2017 Page 74/115
Mabthera Combination Phase: MabThera: 375 mg/m2 + CVP every 21 days
for 8 cycles; number randomized = 231
Endpoints and definitions
Primary endpoint
ORR ORR defined as the proportion of patients
whose best overall disease response (BOR)
either CR or PR during the combination
treatment period (according Central Blinded
Review of radiological response).
Secondary PFS Time from randomisation to either PD or death
Secondary OS Time from randomisation to death
Database lock data cutoff (10-Jul-2015)
Results and Analysis
Analysis description Primary Analysis
Analysis population and time point description
Per Protocol Population At the end of the combination phase
Descriptive statistics and estimate
variability
Treatment group Rixathon MabThera
Number of subject
311 313
ORR 271/311 (87.1%) 274/313 (87.5%)
90%-CI
(83.59%, 90.15%) (84.04%, 90.49%)
Effect estimate per
comparison
Primary endpoint Comparison groups Rixathon - Mabthera
Difference in incidences -0.40%
95%-CI (-5.94, 5.14)
Equivalence margin (-12%, 12%)
Notes Equivalence has been shown. The primary analyses results are supported by
the outcome of various sensitivity analyses: e.g. same analysis excluding patients with missing values: difference in ORR -0.55 (95%-CI: -4.03, 5.14); FAS analysis: difference -0.44 (95%-CI:-5.95%, 5.07%).
Analysis description Secondary analysis
Analysis population and time point
description
Full Analysis Set Start of treatment until date of data cutoff
Descriptive statistics and estimate
variability
Treatment group Rixathon MabThera
Number of
subject
312 315
PFS
(median months)
NE NE
OS (median months)
NE NE
Assessment report
EMA/303207/2017 Page 75/115
Effect estimate per comparison
PFS Comparison groups Rixathon - Mabthera
HR1 1.33
90%-CI (0.98, 1.80)
Effect estimate per comparison
OS Comparison groups Rixathon - Mabthera
HR1 1.03
90%-CI (0.59, 1.80)
Analysis performed across trials (pooled analyses and meta-analysis)
N/A
Clinical studies in special populations
Table 37 - Clinical efficacy in patients aged >65 years old
Age 65 – 74 (Older
subjects number / total
number)
Age 75 – 84 (Older
subjects number / total
number)
Age 85+ (Older
subjects number / total
number)
Rixathon Originator* Rixathon Originator
* Rixathon Originator
*
Controlled Trials
GP13-201 22/133 35/179 6/133 3/179 -- --
GP13-301 63/314 65/315 23/314 15/315 -- --
Non Controlled
Trials
GP13-101 1/6 -- -- -- -- --
*Originator is for GP13-201 study MabThera (data from Part I Week 52) and Rituxan (data from Part II Week 24) combined; for GP13-301 study originator is MabThera.
Supportive study GP13-201
Supportive efficacy data were obtained from study GP13-201, the pivotal PK/PD study conducted in
adult patients with active RA (3rd line treatment- see also Clinical pharmacology section)
The study consisted of a screening period (running from Visit 1 up to but not including Visit 3), an
optional washout period between Visit 1 and Visit 2 for anti-TNF or DMARD, a baseline visit (Visit 2) to
reassess the RA status and associated laboratory testing in order to confirm eligibility of the patient
and to perform a complete disease activity assessment. Visit 3 (Day 1) consisted of randomisation and
first study drug administration.
On Week 24 after the first infusion, patients who were considered responders (defined as having a
decrease in Disease Activity Score (DAS28) and if they had at least residual active disease (DAS28 ≥
2.6) were re-treated with a second course of treatment (either 1000 mg i.v. Rixathon or MabThera on
two separate occasions, 2 weeks apart) up to Week 52 at the discretion of the investigator. Patients
who received this second course of treatment had a safety, efficacy and PD assessment performed 26
weeks after the first infusion of the second course of study medication.
The primary objective was to assess the PK bioequivalence between Rixathon and MabThera in
combination with MTX. Secondary objectives also included several efficacy parameters:
Assessment report
EMA/303207/2017 Page 76/115
Change from baseline in DAS28 (CRP) at Week 24 (key secondary efficacy endpoint) Averaged change from baseline in DAS28 (CRP) between Weeks 4 and 24
DAS28 by visit Change from baseline in DAS28 across study drug batches ACR20 response analysis
Difference in ACR-N scores at Week 24 ACR20, ACR50, ACR70 response analysis EULAR response based on DAS28 (CRP) Disease activity according to DAS28 Disease activity according to SDAI/CDAI Quality of Life Rheumatoid factor (RF) and Anti-CCP antibodies (ACPA)
Efficacy Results Study GP13-201
A total of, 173 patients were randomised, of which 86 patients received Rixathon and 87 patients
received MabThera. The majority of randomised patients (82.1%) completed up to 52 weeks. About
65% of the subjects were re-treated in period following Week 24. In total 6 patients were excluded
from the per-protocol set, because of major protocol violations (not meeting inclusion criteria,
receiving wrong study drug and missing CRP measurements).
Baseline demographics and disease severity were comparable for the Rixathon and MabThera
treatment groups, and reflected the intended target population. The majority of patients were female
(86.1%) and Caucasian (80.9%). The mean age of all patients was 53.7 years and 20.8% of patients
were ≥65 years of age. The mean duration of RA was 10.07 years (range: 1.0 to 34.0 years). All
patients had previously been treated with 1-3 anti-TNFs or other biologic DMARDs, in line with the
protocol inclusion criteria. The mean DAS28 (CRP) scores at baseline were 5.81 (SD=0.92) and 5.85
(SD=0.88), in the Rixathon and MabThera groups, respectively, indicating moderate-severe disease
activity.
In both treatment groups, the DAS28 scores improved significantly form baseline (see Figure 21
below). The mean changes from baseline were comparable between Rixathon and MabThera. The
upper limit of the 95% CI of the difference between both groups was 0.462, below the pre-defined
non-inferiority criterion of 0.6.
Table 38 - Averaged change from baseline in DAS28 (CRP) at Week 24 (PP Analysis Set)
Assessment report
EMA/303207/2017 Page 77/115
Figure 21 - Mean (SD) DAS28-CRP scores, Study GP13-201
The ACR20 (CRP) response rates at Week 24 were 71.8% and 72.4% in the Rixathon and MabThera
group, respectively. The lower limit of the 95% CI for the difference in response rates was -14.74%,
above the pre-defined equivalence margin of -15.0%.
Table 39 - ACR20 responder rates, Week 24, Study GP13-201
Retreatment (i.e. a second treatment course), occurred in about 65% of the participants (59 in
Rixathon arm, and 60 in the MabThera arm), after on average 232.8 (SD62.1) days and 241.5
(SD64.4) days, respectively. Beyond B-cells levels and ADA (anti-drug-antibody), no clinical outcomes
were reported after retreatment.
Table 40 - Summary of Study GP13-201 efficacy in RA
Title: A randomized, double-blind, controlled study to evaluate pharmacokinetics, pharmacodynamics, safety and efficacy of Rixathon and rituximab in patients with rheumatoid arthritis refractory or
intolerant to standard DMARDs and one or up to three anti-TNF therapies.
Study
identifier GP13-201
Design This is a 52-week multicenter, randomized, double-blind, study designed to assess the
pharmacokinetics, pharmacodynamics, safety and efficacy of Rixathon and MabThera® in patients with RA refractory or intolerant to one or up to three anti-TNF therapies.
Study Part I design: Patients received their first course of treatment (study drug infusion
on two separate occasions, two weeks apart) and were followed for 52 weeks. After Week 24, responders (defined as having a decrease in DAS28 derived either with ESR or
CRP of > 1.2 from baseline) could be re-treated with study medication at the discretion of the investigator, if they had at least residual active disease (DAS28 ≥ 2.6).
Duration of main phase: 52 weeks
Assessment report
EMA/303207/2017 Page 78/115
Duration of Run-in phase: None
Duration of Extension phase: None
Hypothesis Equivalence:
Primary objective: Bioequivalence between Rixathon and MabThera (PK)
Secondary objective:
Non-inferiority of Rixathon to MabThera with respect to:
Change from baseline in DAS28 (CRP) disease activity score at Week 24 (using
a mixed model for repeated measures). The upper bound of the 95% confidence limit for the mean difference between Rixathon and MabThera was to be equal or less than 0.6 in order to conclude equivalence.
ACR20 (CRP) response rate at Week 24 (estimated based on pooled standard
error and chi-square statistics). The lower bound of the 95% confident limit was to be greater than -15% in order to conclude equivalence.
Treatments
groups
Rixathon
Rixathon: Two 1000 mg iv infusions, on Day 1 and
Day 15, add-on to methotrexate
Number of patients randomized: 86
MabThera
MabThera: Two 1000 mg iv infusions, on Day 1 and
Day 15, add-on to methotrexate
Number of patients randomized: 87
Efficacy
secondary endpoints:
Equivalence of
Rixathon to MabThera with
respect to change from baseline in DAS28 (CRP) at Week 24
DAS28 (CRP)
at Week 24
Change from baseline in DAS28 (CRP) at Week 24.
A two-sided 95% CI for the mean difference between Rixathon and MabThera® was derived and compared to the pre-specified equivalence margin of 0.6.
Equivalence of
Rixathon to MabThera with respect to ACR20 (CRP) response
rate at Week 24: Comparison between Rixathon and MabThera
ACR20 (CRP)
at Week 24
Percentage of ACR20 responders at Week 24
Database
lock 11-Dec-2014
Results and Analysis
Secondary endpoint
Change from baseline in
DAS28 (CRP) at Week 24
(PPS)
Comparison groups Rixathon vs. MabThera
LSM difference 0.07
SE
95% CI of mean difference
0.201
[-0.328, 0.462]
P-value Not evaluated
Secondary endpoint
ACR20 (CRP) at Week 24
(PPS)
Comparison groups Rixathon vs. MabThera
Response rate difference (%)
-0.57
Assessment report
EMA/303207/2017 Page 79/115
SE
95% CI of mean
difference
7.23
[-14.74, 13.60]
P-value Not evaluated
Analysis description Analysis of key secondary variables:
For the key secondary efficacy endpoint change from baseline in DAS28 at Week
24 (using CRP for the calculation), a mixed model for repeated measures was used, to estimate both mean change from baseline at Week 24 and the averaged change between Week 4 and 24. Two-sided 95% Cl for the mean difference between Rixathon and MabThera were derived and compared to the pre-specified margin of 0.6. The upper limit of the CI was required to be equal to or less than 0.6 in order to claim equivalence.
Analysis of other secondary variables:
A two-sided 95% CI for the difference in both the ACR20 (CRP) response rates
at Week 24 was estimated based on the pooled standard error and chi-square statistic. The lower bound of the confidence limit must be greater than -0.15 or
-15.0% in order to conclude equivalence.
2.5.3. Discussion on clinical efficacy
Design and conduct of clinical studies
Dose response studies are not required for biosimilar products. For the justification of the dose, a
reference is made to the dossier of the originator.
In the pivotal study GP13-301, the choice of the study population of patients with advanced stage
Follicular lymphoma to demonstrate therapeutic equivalence is supported as this is the most common
of the approved oncology indications of MabThera. Moreover, the additive treatment effect to a
background therapy of CVP is expected to be reasonable large in this population based on historical
data, indicating that FL is a sensitive model to detect differences between the biosimilar and originator
product.
The parallel, two-arm, randomised study was adequately designed. The ORR, determined by central
reading, is considered a relevant primary endpoint for this indication. The equivalence margin of +/-
12% is considered acceptable from a clinical perspective. This margin is below the 95% CI of the
treatment effect of the main study of the reference product in this indication. In the placebo-controlled
study of the originator, the ORR was 81% for the MabThera-CVP arm and 57% for CVP + placebo arm,
with a treatment effect of 24% (95% CI: 14%, 34%). Moreover, the choice of 12% margin is also
supported by the indirect confidence interval approach, which was calculated to be 13.4% by the
Applicant, in accordance the EMA Guidance Choice of Non-inferiority Margin. The missing data
imputation of the ORR measures by considering missing data as non-responders, is considered
appropriate.
Efficacy was a secondary endpoint in Study GP13-201 in RA patients, which was powered and designed
to establish bio-equivalence of PK as primary objective. As the study was not powered to establish
non-inferiority (or equivalence) of efficacy, the data on therapeutic endpoints should be considered as
supportive.
In RA, the regular maintenance dosing schedule is every 24 weeks (6 months). Treatment continuation
with a second course of the study drugs after 24 weeks and the timing of a second course was to be
individually decided by the prescriber, based on the level of residual disease activity. Therefore, the
Assessment report
EMA/303207/2017 Page 80/115
data sampling from the second part of the study was more variable, hampering firm conclusions
regarding long-term efficacy after Week 24. The decision making regarding re-treatment was in line
with the posology of MabThera and SAWP recommendations.
The choice of mean change from baseline of DAS28 as ‘key’ secondary endpoints, instead of ACR20
responder rate as originally proposed in the Scientific Advice in 2011, is supported, as continuous
endpoints may be more sensitive to evaluate therapeutic equivalence than dichotomous ones. The NI
margin of 0.6 points is agreed, as it is based on clinical grounds, i.e. a change of DAS28 score of 0.6 is
considered a minimal clinically relevant change. For another secondary endpoint, ACR20, the NI
margin was set at +/- 15%. The cut-off of 15% was based on meta-analysis of other RA studies
showing a difference of ACR20 of 31% (95% CI 25%, 38%) between rituximab and MTX. The 15%
equivalence limit is approximately half of this effect size, and is below the lower limit of the 95% CI.
The 15% margin is considered relatively high from a clinical perspective. However, it is acceptable in
this context. No imputation of missing data was applied for the efficacy outcomes, and the analyses
were performed in the Per-Protocol set. Considering the exploratory character of the study regarding
the clinical efficacy endpoints and that the study was not powered for establishing equivalence of
efficacy in RA, this is accepted.
A high frequency of minor protocol deviations was reported for both studies; most frequently reported
minor protocol deviations being “PK sample out of visit window or missing” (96.5%), “scheduled visits
out of the defined time window” (59.5%), “blood pressure measurement not performed per protocol”
(18.5%), and “premedication not administered per protocol” (16.8%). It is not expected that critical
findings of GCP Inspection of one site would have largely influenced the study outcomes, given the
small number of subject involved, and that three other sites were considered GCP-compliant.
Efficacy data and additional analyses
GP13-301 (pivotal efficacy trial)
The demographics were largely similar between the treatment arms in terms of age, gender, race, or
other demographic parameters. Patients treated with Rixathon were slightly more likely older, more
female, Caucasian (and not Asian), with better ECOG but only with minor differences. Regarding
disease history there is a slight increase of patients with bulky or stage 4 disease in the Mabthera
treated group.
The ORR was 87.1% in the Rixathon arm and 87.5% in the MabThera arm and the difference in ORRs
was -0.40% (95% CI [-5.94%, 5.14%]; 90% CI [-5.10%, 4.30%]). Equivalence was concluded as the
entire 95% CIs for the difference in ORR between the two treatments was within the pre-specified
equivalence margin of ±12%. Thus, the primary endpoint was met. This was further supported by pre-
defined sensitivity analyses.
Subgroup analyses according to age demonstrated similarity. However, the two arms were numerically
different for ORR randomized by FLIPI strata, favoring MabThera (91.2% vs. 82.8% in Rixathon) in the
subset of patients with a FLIPI score 0-2 (low-intermediate risk), and favoring Rixathon (90.4% vs.
84.7%) in the subset of patients with a FLIPI score 3-5 (high risk). However, an analysis of PFS and
OS according to FLIPI score did not reveal major differences with regards to outcome between the
treatment arms.
In general, the proportions of patients with best overall response (BOR) for the four categories (CR,
PR, SD and PD) were similar. The results of the sensitivity analysis repeated using the FAS were
consistent with those seen in the PPS.
Assessment report
EMA/303207/2017 Page 81/115
Notably, the rate of Complete Remission (about 14%) was significant lower in Study 301 than reported
before for other trials (Marcus: 30%, Federico: 67%). This might be explained by differences in the
study populations. Patients in the Study GP13-301 had a higher risk at baseline (FLIPI 3 score -5 in
56% of patients), than in the historical trials (22-38%) The CR rate was assessed and confirmed by
central review, which was not the case with public available CR rates from historical trials where only
CR rates from local reading are available. Thus, the CR outcomes of Study GP13-301 met more
stringent criteria; however, the CR rates were balanced between both treatment arms of Study GP13-
301, and the overall response (ORR) is high and similar. Assay sensitivity could still be assumed
despite the relative low CR rates, as the overall ORR is in the same range as reported before, and
response might be even lower in populations at higher risks –providing a potential larger treatment
effect.
When comparing best overall response based on investigator versus central blinded review it was
observed that for 114 patients who had a CR as per investigator, the central review confirmed only 52
as a CR and 58 as a PR with a few results being determined as SD, PD and unknown. However, this
discrepancy is not viewed as important since (a) primary analyses were conducted under central
review, (b) the discrepancy is only one stage category and (c) even with this difference the primary
endpoint is ORR which is the sum of both CR + PR. However, such discrepant findings may impact on
the PFS analysis, which was conducted as secondary analysis and by investigator review only (see
further below).
PFS was based on investigator assessment. The median PFS was not reached for either of the two
treatment groups. As of the time of data cutoff (10-Jul-2015), respectively 21.5% [n=67] and 16.5%
[n=52] of patients in Rixathon and MabThera treatment arms progressed or died, with 62.5% [n=195]
and 71.7% [n=226] patients censored without event and having adequate follow-up, respectively. The
majority of these patients were still ongoing after the Combination Treatment Phase so their eventual
PFS outcomes in the study may still potentially alter the PFS curve of either treatment arm beyond 6
months.
The PFS HR (Rixathon/MabThera) was 1.33 (90% CI: [0.98, 1.80]), based on the investigator
assessment indicating that PFS may deviate between assignments in the treatment maintenance
phase, not in favour of Rixathon (data cut-off 10-Jul-2015). It has to be taken into consideration that
the study is not adequately powered to demonstrate equivalence of PFS. With the PFS update of 10-
Jul-2016 the hazard ratio is 1.25 (with 90% CI: [0.96, 1.61]) thus decreasing with an increase in
number of events over time. The median follow-up time was 18.15 months for the Rixathon group and
22.80 months for the MabThera group. The median PFS could not yet be estimated for either treatment
group based on the currently available data. Considering that the majority of PFS events occurred
during the Maintenance Phase, these figures are still suggestive of immaturity of PFS data. Sensitivity
analyses were performed on the PFS in the ongoing monotherapy maintenance treatment phase. The
landmark analyses, selecting patients who were at risk of progression at the time of entrance of the
maintenance phase –i.e. those patients without progression in the first combination-therapy study
phase-, showed similar outcomes as the ITT FAS analyses. Refined criteria for PFS events and
censoring did not lead to different conclusions. When exploring differences of time-interference for
either treatment, using Kaplan-Meier estimates up to the time point of the last event occurred, a
statistically significant difference in PFS was observed after 24 months between Rixathon and
MabThera, but no statistically significant differences were noted at other time-points. However, Month
24 is only covering a third of the enrolled study population, and the data were too immature for final
conclusions on differences in time-interference. In the updated submission (data cut off: 31-Dec-2016)
the number of PFS events was still low and the rate of censoring high. The divergence between the two
KM curves has not decreased with these updated data. The PFS HR (Rixathon/MabThera) was 1.31
Assessment report
EMA/303207/2017 Page 82/115
(90% CI [1.02, 1.69]), in the same range as observed with the first PFS analysis (data cut off: 10-Jul-
2016) thus in line with the PFS findings from the 1st PFS interim analysis. The follow-up time is still too
short to allow for an estimation of median PFS.
The divergence is considered due to patient heterogeneity or random data variation rather than a real
treatment effect; the study was not powered to demonstrate similarity (nor to detect a difference) in
PFS between the products, and that the PFS results should be interpreted with caution. Moreover, the
potential PFS difference is not reflected by an effect in CR rates at various time points (month 15, 27,
33 and at end of study). CR has been shown to correlate with OS (even as surrogate marker when
measured at week 30) whereas PFS has not. [Shi Q, Flowers CR, Hiddemann W et al (2017); Thirty-
months complete response as a surrogate end point in first-line follicular lymphoma therapy: An
individual patient-level analysis of multiple randomized trials. J Clin Oncol 35(5): 552-560].
Furthermore, the interpretability of the PFS results is hampered by the study design (PFS assessment
at only 6 month time interval, and no planned assessment for disease progression beyond 3 years
follow up), the immaturity of the data, the high level of censoring (70% of the patients with main
reason for censoring “adequate assessment no longer available”, ~50% of censored patients), and
median follow up time of less than 2 years.
As of July 2015, the number of deaths (18 events [5.8%] in the Rixathon group vs. 17 events [5.4%]
in the MabThera group) were similar between the two treatment arms. The median OS has not yet
been reached for either treatment group based on the currently available data. The analysis of OS
using the Cox regression method showed that the HR was 1.03 (90% CI: 0.59, 1.80) at July 2015 and
with the data update from 15-Jul-2016 the HR for OS changed to 0.78 and to 0.77 (90% CI: [0.49,
1.22]) at data cut off: 31-Dec-2016. The analysis of overall survival using the Kaplan-Meier and Cox-
regression method for the PPS was consistent with the findings for the FAS. The currently observed
data are still premature and cannot meaningfully contribute to the biosimilarity assessment. Updated
PFS and OS data are to be expected with the next planned update (see RMP).
GP13-201 (efficacy data in RA)
The key secondary efficacy endpoint was change from baseline in DAS28 (CRP) at Week 24. The least
square (LS) mean difference between Rixathon and MabThera in the change from baseline in DAS28
(CRP) at Week 24 was 0.07 (95% CI -0.328, 0.462). The upper limit of the corresponding 95% CI was
0.462, which is below the pre-defined non-inferiority margin of 0.6. Thus, the criterion for non-
inferiority was met.
The mean difference between Rixathon and MabThera in the averaged change from baseline in DAS28
(CRP) between Weeks 4 and 24 was 0.33 (95% CI 0.029, 0.639). The upper limit of the corresponding
95% CI was 0.639, slightly above the pre-defined non-inferiority margin of 0.6; indicating a marginal
difference compared to MabThera. However, the results do not indicate clinically relevant differences
between both treatment groups. Analysis of DAS 28 per visit by treatment group suggests similarity.
Mean DAS28 (CRP) was numerically lower for the MabThera treatment arm compared to Rixathon until
Week 24. After Week 24, a reverse trend was observed up to Week 52. Both treatment arms,
however, exhibit a large degree of variation at all timepoints as shown by the wide SD markers. Post-
hoc analysis showed no clear trend for DAS28 with regards to Cmax, suggesting that the observed
variability in Cmax had no influence on efficacy in either of the treatment arms.
The mean profiles for both Rixathon batches lie at the upper end, meaning less mean change of DAS
28 from baseline, consistent with the observation in the primary analysis of DAS28 (change from
week 4 to 24) and more pronounced in the average change from week 4 to 24. However, both
Rixathon batches lie within the range of mean profiles for different MabThera batches. For all batches,
there is a degree of variability in the mean change from baseline in DAS28 (CRP) profiles.
Assessment report
EMA/303207/2017 Page 83/115
The ACR20 (CRP) response rates at Week 24 were similar for the two treatment groups (71.8% and
72.4% in the Rixathon and MabThera groups, respectively). The lower bound of the 95% CI for the
difference in response rates was -14.74%, above the pre-defined non-inferiority margin of -15.0%.
Further analyses for ACR20 (CRP) were conducted (averaged ACR20 (CRP) responder estimate
between Week 4 and 24; averaged ACR20 (CRP) at week 24). All analyses show that MabThera and
Rixathon are similar regarding ACR20 rates with when applying equivalence margins of ±15%.
The difference between ACR20 (CRP) scores at different time points -except for weeks 16 and 38-, with
a confidence interval contained within +/- equivalence margins, were discussed but were not viewed as
clinically meaningful. In addition, slight differences were observed up to and including week 16
(ACR50, ACR70) or up to and including week 24 (SDAI/CDAI). The differences in efficacy scores/values
between Rixathon and Mabthera among weeks 4 to 24 are decreasing with the data update (cut-off
10-Jul-2015) and were not supported by the comparisons to the US product Rituxan to which bridging
of data has been made.
In overall, results from the ACR20 response analysis at week 24 and averaged ACR20 responders
between week 4 and week 24 showed that Rixathon and MabThera are similar regarding ACR20 rates
with when applying equivalence margins of ±15%.in study GP13-201 the criterion for non-inferiority in
terms of the key secondary efficacy endpoint change from baseline in DAS28 (CRP) at Week 24- was
met.
Demonstrating equivalence of efficacy in both a model of the haematology-oncology and for the RA
indication, the most common of the immunology indications, supports the extrapolation to the non-
studied oncology and immunology indications, together with the evidence from functional assays and
PK-PD studies.
2.5.4. Conclusions on the clinical efficacy
Efficacy and biosimilarity to Mabthera has been demonstrated in an adequately designed randomised
study, 627 patients with follicular non-Hodgkin lymphoma (FL) as regards ORR (supported by
sensitivity analyses in the FAS population) and BOR. Efficacy and biosimilarity to Mabthera has been
shown in a supportive study in patients with RA.
The efficacy outcomes of the confirmatory study in FL and the exploratory study in RA support
extrapolation to the other non-studied oncology and immunology indications.
Assessment report
EMA/303207/2017 Page 84/115
2.6. Clinical safety
Table 41 - Safety data set
Patient exposure
A total of 404 patients (86 RA patients and 318 NHL patients; 87.1 Patients’Years (PY)) have been
exposed to Rixathon in studies GP13-201, GP13-301 and GP13-101.
Prolonged treatment continuation of Rixathon up to a maximum 2.5 years was applied in 231 patients
with Follicular Lymphoma. The maintenance treatment part of Study GP13-301 is still ongoing. At the
cut-off date of data sampling, about 50% of the study population had received at least 4 cycles of
Rixathon/MabThera. Total exposure to Rixathon was 476.4 PY at cut-off of 10 July 2015.
Assessment report
EMA/303207/2017 Page 85/115
Table 42 - Study drug administration and compliance, by treatment in patients with RA – GP13-201 (Safety Analysis Set)
Table 43 - Exposure to investigational study treatment in patients with FL by cumulative
dose, dose intensity, and relative dose intensity – GP13-301 Combination and Maintenance Phase (Safety Set)
Pooling of safety data of these studies is not considered meaningful, considering the differences in
rituximab standard treatment regimen between oncology and auto-immune indications, which is by far
more intensive for NHL than for RA. Moreover, there are differences in background risks of the
patients’ populations and their concurrent therapies, which chemotherapy in NHL and methotrexate in
RA. Therefore, the safety outcomes are discussed separately for each clinical study in this report.
Assessment report
EMA/303207/2017 Page 86/115
Adverse events
GP13-201
The overall incidence of AEs was comparable for the two treatment groups (Rixathon: 56 patients,
65.1% vs. MabThera: 57 patients, 65.5%) with no clinically meaningful differences between the
treatment groups for any SOC.
The most commonly affected primary SOC was infections and infestations (33.5% overall). This is in
line with earlier studies with MabThera. The next 2 most commonly affected primary SOCs were
musculoskeletal and connective tissue disorders (17.3% overall) and gastrointestinal disorders (16.2%
overall).
The most common AEs overall included urinary tract infection, nasopharyngitis, (worsening of)
rheumatoid arthritis, hypertension, upper respiratory tract infection, and bronchitis. Urinary tract
infection occurred more often in the Rixathon group (10.5%) compared with the MabThera group
(5.7%). None of the AEs of urinary tract infection in the Rixathon group were serious, all were mild in
severity (except for one moderate), and all resolved. This was further addressed in the response to
Q89. There were otherwise no clinically meaningful differences between the treatment groups for any
of the most frequently occurring AEs.
Table 44 - Summary of AE categories in patients with RA – GP13-201 (Safety Analysis Set)
AEs suspected to be related to study drug based on investigator assessment occurred equally between
the two treatment groups (Rixathon 28 patients; 32.6%, MabThera 29 patients, 33.3%).
The most common suspected drug-related AEs were related to infections and infestations with
comparable incidence between the two treatment groups (Rixathon 13 patients, 15.1%; MabThera 15
patients, 17.2%). The next most common AEs were related to Respiratory, thoracic and mediastinal
disorders (Rixathon 2 patients, 2.3%; MabThera 7 patients, 8%), vascular disorders (Rixathon 4
patients, 4.7%; MabThera 4 patients, 4.6%), gastrointestinal disorders (Rixathon 2 patients, 2.3%;
MabThera 6 patients, 6.9%) and general disorders and administrative site conditions (Rixathon seven
patients, 8.1%; MabThera one patient, 1.1%).
Assessment report
EMA/303207/2017 Page 87/115
The five most common AEs rated to have a relationship to study drug overall were urinary tract
infection (4%), nasopharyngitis (3.5%), hypertension (2.9%), infusion related reaction (2.9%) and
pruritus (2.9%). There was a higher rate of urinary tract infections for Rixathon in comparison to
MabThera (10.5 versus 5.7%). In contrast, the overall rates of pruritus and infections were slightly
higher for MabThera. For the latter this was attributed to a higher rate of respiratory tract infections
(bronchitis: Rixathon 3.5%, MabThera 5.7%; upper respiratory tract infection: Rixathon 3.5%,
MabThera 5.7%; cough: Rixathon 2.3%, MabThera 4.6%; respiratory tract infection: Rixathon none,
MabThera 3.4%).
GP13-301
Combination Phase
The rate and severity of adverse events was considerable higher for FL versus RA patients. This is to
be expected considering the more frequent rituximab treatment regimen of 8 dosing cycles within 6
months, as compared to a 24 weeks dosing regimen in RA, and the combination with CVP in FL
treatment.
The overall incidence of AEs during the combination phase was similar in both treatment groups
(Rixathon-CVP: 92.6%; MabThera-CVP 91.4%). The most commonly (>30% incidence in either
treatment group) affected primary SOCs were gastrointestinal disorders (primarily constipation and
nausea), nervous system disorders (primarily peripheral neuropathy, and paraesthesia) and infections
and infestations (primarily urinary tract infection and upper respiratory tract infection).
Different than in the RA study, neutropenia, peripheral neuropathy/paraesthesia, fatigue and
nausea/constipation or abdominal pain were frequently reported in the FL study population, with
frequencies over 10%. In comparison, in the RA study only one case of neutropenia was reported (in
the MabThera arm), and a low frequency of fatigue (2.4%), abdominal pain (1.7%) and paraesthesia
(1.2%). These differences are probably due to CVP co-treatment or other co-medications (like opioids),
and/or differences in the rituximab treatment schedule, or the disease itself.
The only AE reported with at least a 5% absolute difference and more frequently in the Rixathon group
than in the MabThera group was peripheral neuropathy (15.1% vs. 9.5%, respectively). The only AE
reported with at least a 5% absolute difference and occurring more frequently in the MabThera group
compared to the Rixathon group was paraesthesia (14.3% vs. 8.3%), although it must be noted that
peripheral neuropathy and paresthesia are essentially the same and taken together, the combined
incidence rate of peripheral neuropathy and paraesthesia is similar between the treatment groups (i.e.,
Rixathon with 15.1% + 8.3% = 23.4% vs. MabThera with 9.5% + 14.3% = 23.8%).
The incidence of grade 3 and 4 AEs were similar between the two treatment groups (Rixathon grade 3
40.7%, grade 4 12.5%; MabThera grade 3 41.9%, grade 4 14.9%).
Assessment report
EMA/303207/2017 Page 88/115
Table 45 - Summary of AE categories in patients with FL – GP13-301 Combination Phase (Safety Set)
In the Combination Phase, AEs considered as study drug-related were reported by 230 (73.7%)
Rixathon and 223 (70.8%) MabThera patients. These events were primarily grade 1 (mild) or 2
(moderate). The most common AEs were neutropenia, constipation and infusion-related reactions.
Maintenance Phase
During the Maintenance Phase, AEs were reported by 63.2% in the Rixathon group and by 57.1% in
the MabThera group. With the data update this percentage difference decreased from 6.1% to 1.6%.
The most commonly affected (>15% incidence in either treatment group) primary SOCs were
infections and infestations (primarily upper respiratory tract infection and urinary tract infection),
musculoskeletal and connective tissue disorders (primarily arthralgia and back pain), gastrointestinal
disorders (primarily diarrhea and vomiting), and general disorders and administration site conditions
(primarily asthenia and pyrexia). Six AEs occurred with a frequency of >5% in either treatment group.
While neutropenia and cough occurred more frequently in the Rixathon group, the other common AEs
occurred more frequently in the MabThera group.
The incidence of grade 3 AEs were similar between the two treatment groups (Rixathon 13.9%;
MabThera 13.0%), while there were more grade 4 AEs occurring in the Rixathon group (12 patients,
5.2%) compared to the MabThera group (4 patients, 1.7%). Neutropenia was the only AE that was
reported to occur at a greater than 2% frequency (Rixathon, grade 3/4 in 17 patients [7.4%]
MabThera, grade 3/4 in nine patients [3.9%]). However, this difference was not accompanied by a
higher number of febrile neutropenia or infections (SOC infections and infestations - Rixathon: 32.7%;
Mabthera: 36.5%, cut-off 10-Jul-2016).
Assessment report
EMA/303207/2017 Page 89/115
Table 46 - Summary of AE categories in patients with FL – GP13-301 Maintenance Phase (Maintenance Set)
Category
GP2013
N=231 n(%)
MabThera
N=231 n(%)
Adverse events (AEs) 146 (63.2) 132 (57.1)
Suspected to be drug-related 54 (23.4) 38 (16.5)
Grade 3-4 AEs 39 (16.9) 32 (13.9)
Suspected to be drug-related 16 (6.9) 12 (5.2)
Deaths 2 (0.9) 2 (0.9)
Serious adverse events (SAEs) 14 (6.1) 10 (4.3)
Suspected to be drug-related 6 (2.6) 3 (1.3)
AEs leading to discontinuation 8 (3.5) 4 (1.7)
Potential infusion related reaction 85 (36.8) 88 (38.1)
Suspected to be drug-related 25 (10.8) 19 (8.2)
AEs requiring dose interruption and/or reduction 17 (7.4) 17 (7.4)
AEs requiring additional therapy1 111 (48.1) 104 (45.0)
In the Maintenance Phase, AEs considered as study drug-related were reported by 54 (23.4%)
Rixathon and 38 (16.5%) MabThera patients. These events were primarily grade 1 (mild) or 2
(moderate).
There were 16 (6.9%) Rixathon and 12 (5.2%) MabThera patients reporting drug-related grade 3/4
AEs.
Serious adverse event/deaths/other significant events
GP13-201
Overall, the number of patients with SAEs was comparable for the two treatment groups with no
clinically meaningful difference between the treatment groups (Rixathon: 10 patients (11.6%) vs.
MabThera: 14 patients (16.1%)). SAEs were reported most often in the infections and infestations
SOC: 5 (5.8%) patients in the Rixathon group and 4 (4.6%) patients in the MabThera group.
SAEs judged by the investigator to have a relationship to study drug were reported overall in 3 (3.5%)
patients in the Rixathon group and 6 (6.9%) patients in the MabThera group. The most common of
these were reported in the infections and infestations SOC: 3 (3.5%) patients in the Rixathon group
(abscess, groin abscess, Klebsiella sepsis, and soft tissue infection) and 2 (2.3%) patients in the
MabThera group (atypical pneumonia and pneumonia haemophilus).
The proportions of patients with other non-fatal SAEs or who discontinued study treatment due to AEs
were generally comparable for the two treatment groups.
One patient in the Rixathon group died during the study due to a treatment-emergent adverse event of
multi-organ failure after an accidental overdose of MTX (daily dose instead of weekly intake). As this
event occurred 18 weeks after the first course of treatment, causality seems unlikely.
Assessment report
EMA/303207/2017 Page 90/115
GP13-301
Combination Phase
During the Combination Phase, the incidence of SAEs was similar between the Rixathon (71 patients,
22.8%) and MabThera (63 patients, 20.0%) groups.
In the SOC of blood and lymphatic disorders, SAEs were reported for 23 (7.4%) Rixathon and 17
(5.4%) MabThera patients. The SAE of infusion-related reaction occurred in three Rixathon patients
and one MabThera patients. For infections and infestations, the number of patients with an SAE was
equal (21 patients each, 6.7%) for both treatment groups.
The incidence of individual SAEs was <5% in either treatment group; the majority of reported SAEs
occurred in one or two patients. The most common SAEs were febrile neutropenia (Rixathon 4.8%;
MabThera 2.9%), pyrexia (Rixathon 1.3%; MabThera 2.2%), abdominal pain (Rixathon 1.3%;
MabThera 1.9%), neutropenia (Rixathon 1.3%; MabThera 1.6%), and sepsis (Rixathon 0.6%;
MabThera 1.6%).
The incidence of study drug-related SAEs was similar between the Rixathon (32 patients, 10.3%) and
MabThera (25 patients, 7.9%) groups. The incidence of specific SAEs was <4% for either treatment
group. The most common related SAE was febrile neutropenia (Rixathon 11 patients, 3.5%; MabThera
9 patients, 2.9%). Differences between treatment groups for specific SAEs were <1%.
Maintenance Phase
During the Maintenance Phase, the incidence of SAEs was low and slightly higher for Rixathon (14
patients, 6.1%) than for MabThera (10 patients, 4.3%). This difference also decreased with the data
update from 10-Jul-2016 (20 [7.9%] for Rixathon and 18 [7.1%] for MabThera). The incidence of
individual SAEs was <1% for both treatment groups. The types of SAEs reported were similar between
the two treatment groups. The SOC with the highest reported number of SAEs was infections and
infestations (7 patients Rixathon; 6 patients MabThera). The causes of SAE were very diverse, and no
clear pattern emerged
• Deaths
Mortality rates were similar between both treatments throughout Study GP13-301. During the entire
study for all phases combined (Combination, Maintenance, and Posttreatment), 35 (5.6%) patients
died (18 [5.8%] in the Rixathon group and 17 [5.4%] in the MabThera group). The most common
cause of death during the study was Non- Hodgkin's lymphoma in both treatment arms, eight (2.6%)
in the Rixathon group and six (1.9%) in the MabThera group.
During the Maintenance Phase, 4 patients died (2 in the Rixathon group and 2 in the MabThera group
These cases were not considered related to rituximab treatment.
Adverse events of special interest
GP13-201
The overall incidence of AEs considered to be of special interest was comparable for the two treatment
groups (24.4% vs. 25.3%) with no clinically meaningful differences between the treatment groups for
any SOC. The two most commonly reported AEs of special interest were rheumatoid arthritis (Rixathon
4 patients, 4.7%; MabThera 5 patients, 5.7%) and hypertension (Rixathon 3 patients, 3.5%;
MabThera 5 patients, 5.7%).
Assessment report
EMA/303207/2017 Page 91/115
• Infusion-related reactions
The numbers of patients who reported at least one IRR at any time during the observation period
overall was 32 (37.2%) patients in the Rixathon group and 37 (42.5%) patients in the MabThera
group. The most common of these IRRs were hypertension (4.6% overall), back pain (4% overall),
headache (4% overall), nausea (4% overall), and pruritus (4% overall).
The incidence of potential infusion related reactions with suspected causal relationship to the study
drug was similar between both treatment groups (Rixathon: 18 patients, 20.9%; MabThera: 19
patients, 21.8%).
Infusion related reactions occurring on the day or the day after the infusion were considered to be
potentially hyper-acute or acute infusion reactions. After the first infusion, such events occurred in a
total of 24 (13.9%) patients (Rixathon: 10 patients, 11.6%; MabThera: 14 patients, 16.1%). After the
second infusion, fewer events of potential infusion related reactions occurred on the day or the day
after the infusion, i.e. in a total of 11 (6.4%) patients (Rixathon: 4 patients, 4.7%; MabThera: 7
patients, 8.0%). Despite the numerical differences, no clinically relevant differences were noted
between the groups.
GP13-301
For study GP13-301, cytokine release syndrome and PML were defined as AEs of special interest in the
clinical study protocol. No PML occurred during the study. During Combination Phase, cytokine release
syndrome occurred in 2 (0.6%) patients during the Maintenance Phase.
• Infusion-related reactions
Combination Phase
During the Combination Phase of study GP13-301, the frequency of patients experiencing the AE
“infusion related reaction” with suspected relationship to study drug was similar between both
treatment groups (Rixathon: 13.1%, grade 3 or 4 1.0%; MabThera: 11.7%, grade 3 or 4 0.6%).
The frequency of patients experiencing the AE “infusion related reaction” requiring study drug dose
adjustments or temporary interruptions during the Combination Phase was also similar in both
treatment arms (Rixathon: 6.7%, grade 3 or 4 1.0%; MabThera: 7.0%, grade 3 or 4 0.6%).
One patient in each treatment arm discontinued the treatment due to the AE “infusion related reaction”
during the Combination Phase. SAEs of infusion related reactions with suspected causal relationship to
study drug were reported with a frequency of 1.0 % in patients treated with Rixathon and of 0.3%
treated with MabThera during the Combination Phase.
Maintenance Phase
During the Maintenance Phase of GP13-301, the frequency of potential infusion related reactions with
suspected causal relationship to study treatment was similar between the Rixathon (10.8%) and
MabThera (8.2%) groups.
The most frequently reported AEs related to infusion related reactions (i.e. >1% of patients in either
treatment group) were infusion related reaction (Rixathon: 1.3%; MabThera: 1.3%), cough (Rixathon:
2.2%; MabThera: none), thrombocytopenia (Rixathon: 0.4%; MabThera: 1.3%), and arthralgia
(Rixathon: none; MabThera: 1.3%). Other infusion related reactions were reported by 1 or 2 patients
in either treatment group, with a similar frequency and pattern.
Assessment report
EMA/303207/2017 Page 92/115
Serious adverse events and deaths
GP13-201
Overall, the number of patients with SAEs was comparable for the two treatment groups with no
clinically meaningful difference between the treatment groups (Rixathon: 10 patients (11.6%) vs.
MabThera: 14 patients (16.1%)). SAEs were reported most often in the infections and infestations
SOC: 5 (5.8%) patients in the Rixathon group and 4 (4.6%) patients in the MabThera group.
SAEs judged by the investigator to have a relationship to study drug were reported overall in 3 (3.5%)
patients in the Rixathon group and 6 (6.9%) patients in the MabThera group. The most common of
these were reported in the infections and infestations SOC: 3 (3.5%) patients in the Rixathon group
(abscess, groin abscess, Klebsiella sepsis, and soft tissue infection) and 2 (2.3%) patients in the
MabThera group (atypical pneumonia and pneumonia haemophilus).
The proportions of patients with other non-fatal SAEs or who discontinued study treatment due to AEs
were generally comparable for the two treatment groups.
One patient in the Rixathon group died during the study due to a treatment-emergent adverse event of
multi-organ failure after an accidental overdose of MTX (daily dose instead of weekly intake). As this
event occurred 18 weeks after the first course of treatment, causality seems unlikely.
GP13-301
Combination Phase
During the Combination Phase, the incidence of SAEs was similar between the Rixathon (71 patients,
22.8%) and MabThera (63 patients, 20.0%) groups.
In the SOC of blood and lymphatic disorders, SAEs were reported for 23 (7.4%) Rixathon and 17
(5.4%) MabThera patients. The SAE of infusion-related reaction occurred in three Rixathon patients
and one MabThera patients. For infections and infestations, the number of patients with an SAE was
equal (21 patients each, 6.7%) for both treatment groups.
The incidence of individual SAEs was <5% in either treatment group; the majority of reported SAEs
occurred in one or two patients. The most common SAEs were febrile neutropenia (Rixathon 4.8%;
MabThera 2.9%), pyrexia (Rixathon 1.3%; MabThera 2.2%), abdominal pain (Rixathon 1.3%;
MabThera 1.9%), neutropenia (Rixathon 1.3%; MabThera 1.6%), and sepsis (Rixathon 0.6%;
MabThera 1.6%).
The incidence of study drug-related SAEs was similar between the Rixathon (32 patients, 10.3%) and
MabThera (25 patients, 7.9%) groups. The incidence of specific SAEs was <4% for either treatment
group. The most common related SAE was febrile neutropenia (Rixathon 11 patients, 3.5%; MabThera
9 patients, 2.9%). Differences between treatment groups for specific SAEs were <1%.
Maintenance Phase
During the Maintenance Phase, the incidence of SAEs was low and slightly higher for Rixathon (14
patients, 6.1%) than for MabThera (10 patients, 4.3%). This difference also decreased with the data
update from 10-Jul-2016 (20 [7.9%] for Rixathon and 18 [7.1%] for MabThera). The incidence of
individual SAEs was <1% for both treatment groups. The types of SAEs reported were similar between
the two treatment groups. The SOC with the highest reported number of SAEs was infections and
infestations (7 patients Rixathon; 6 patients MabThera). The causes of SAE were very diverse, and no
clear pattern emerged
Assessment report
EMA/303207/2017 Page 93/115
• Deaths
Mortality rates were similar between both treatments throughout Study GP13-301. During the entire
study for all phases combined (Combination, Maintenance, and Posttreatment), 35 (5.6%) patients
died (18 [5.8%] in the Rixathon group and 17 [5.4%] in the MabThera group). The most common
cause of death during the study was Non- Hodgkin's lymphoma in both treatment arms, eight (2.6%)
in the Rixathon group and six (1.9%) in the MabThera group.
During the Maintenance Phase, 4 patients died (2 in the Rixathon group and 2 in the MabThera group.
These cases were not considered related to rituximab treatment.
Laboratory findings
GP13-201
The incidence of clinically notable and newly occurred hematology values as well as values in clinical
chemistry was low and similar between both treatment groups. The number of patients with increased
serum creatinine was higher in the Rixathon group. No relevant differences in any vital signs were
observed between Rixathon and MabThera groups in study GP13-201. ECG was only performed at
screening, therefore, no analysis is available.
GP13-301
During the Combination the incidence of ECG abnormalities was higher in the Rixathon group (18
patients vs 10 patients in the MabThera group). Abnormalities in clinical chemistry were similar for
both treatment groups, including abnormal values of CTCAE grade 3 or 4. No relevant differences in
any vital signs were observed between Rixathon and MabThera groups.
In the monotherapy maintenance phase, the rate of haematological abnormalities, that worsened or
newly emerged from the start of the maintenance phase (i.e. after finishing the combination with CVP
phase), was similar for both treatments
Safety in special populations
Table 47 - Safety according to age groups (NHL)
MedDRA Terms Age < 65
(number
(percentage)
N=5
Age 65 – 74
number
(percentage)
N=1
Age 75 – 84
number
(percentage)
N=0
Age 85+
number
(percentage)
N=0
Total AEs 4 (80%) 1 (100%) -- --
Serious AEs - Total 0 (0%) 0 (0%) -- --
- Fatal -- -- -- --
- Hospitalization/
prolong existing
hospitalization
-- -- -- --
- Life-threatening -- -- -- --
- Disability/incapacity -- -- -- --
- Other (medically
significant)
-- -- -- --
AE leading to drop-out 0 (0%) 0 (0%) -- --
Psychiatric disorders 0 (0%) 0 (0%) -- --
Assessment report
EMA/303207/2017 Page 94/115
MedDRA Terms Age < 65
(number
(percentage)
N=5
Age 65 – 74
number
(percentage)
N=1
Age 75 – 84
number
(percentage)
N=0
Age 85+
number
(percentage)
N=0
Nervous system
disorders
1 (20%) 0 (0%) -- --
Accidents and injuries 1 (20%) 0 (0%) -- --
Cardiac disorders 0 (0%) 0 (0%) -- --
Vascular disorders 0 (0%) 0 (0%) -- --
Cerebrovascular
disorders
0 (0%) 0 (0%) -- --
Infections and
infestations
1 (20%) 0 (0%) -- --
Anticholinergic
syndrome
0 (0%) 0 (0%) -- --
Quality of life
decreased
Not evaluated Not evaluated -- --
Sum of postural
hypotension, falls,
black outs, syncope,
dizziness, ataxia,
fractures
0 (0%) 0 (0%) -- --
< other AE appearing
more frequently in
older patients>
0 (0%) 0 (0%) -- --
Table 48 - Safety according to age groups in study GP13-201 (Rheumatoid arthritis)
MedDRA Terms Age < 65
(number
(percentage)
Age 65 – 74
number
(percentage)
Age 75 – 84
number
(percentage)
Age 85+
number
(percentage)
Rixatho
n
N=105
Orig.
N=141
Rixatho
n
N=22
Orig.
N=35
Rixatho
n
N=6
Orig.
N=3
Rixatho
n
N=0
Orig.
N=0
Total AEs 67(63.8
)
80(56.7
)
13(59.1
)
25(71.4
)
5 (83.3) 2(66.7
)
-- --
Serious AEs - Total 9 (8.6) 14 (9.9) 3 (13.6) 5 (14.3) 1 (16.7) 0
(0.0)
-- --
- Fatal 1 (1.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0
(0.0)
-- --
- Hospitalization/
prolong existing
hospitalization
7 (6.7) 13 (9.2) 3 (13.6) 4 (11.4) 1 (16.7) 0
(0.0)
-- --
- Life-threatening -- -- -- -- -- -- -- --
- Disability/
incapacity
-- -- -- -- -- -- -- --
- Other
(medically
significant)
-- -- -- -- -- -- -- --
AE leading to drop-
out
2 (1.9) 6 (4.3) 2 (9.1) 2 (5.7) 0 (0.0) 0
(0.0)
-- --
Psychiatric 4 (3.8) 5 (3.5) 0 (0.0) 1 (2.9) 0 (0.0) 0 -- --
Assessment report
EMA/303207/2017 Page 95/115
MedDRA Terms Age < 65
(number
(percentage)
Age 65 – 74
number
(percentage)
Age 75 – 84
number
(percentage)
Age 85+
number
(percentage)
disorders (0.0)
Nervous system
disorders
10 (9.5) 16(11.3
)
3 (13.6) 4 (11.4) 0 (0.0) 0
(0.0)
-- --
Accidents and
injuries
15(14.3
)
12 (8.5) 3 (13.6) 5 (14.3) 1 (16.7) 0
(0.0)
-- --
Cardiac disorders 5 (4.8) 4 (2.8) 0 (0.0) 1 (2.9) 0 (0.0) 0
(0.0)
-- --
Vascular disorders 9 (8.6) 8 (5.7) 1 (4.5) 6 (17.1) 2 (33.3) 0
(0.0)
-- --
Cerebrovascular
disorders
0 (0.0) 1 (0.7) 0 (0.0) 0 (0.0) 0 (0.0) 0
(0.0)
-- --
Infections and
infestations
31(29.5
)
40(28.4
)
9 (40.9) 11(31.4
)
1 (16.7) 1(33.3
)
-- --
Anticholinergic
syndrome
0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0
(0.0)
-- --
Quality of life
decreased
Not evaluated
Sum of postural
hypotension, falls,
black outs,
syncope, dizziness,
ataxia, fractures
9 (8.6) 4 (2.8) 2 (9.1) 3 (8.6) 2 (33.3) 0
(0.0)
-- --
< other AE
appearing more
frequently in older
patients>
-- -- -- -- -- -- -- --
GP13-201: data from GP13-201 Part 1 Week 52 and GP13-201 Part II Week 24 data combined.
The effect of intrinsic and extrinsic factors was not evaluated.
Immunological events
Immunogenicity was evaluated in terms of antibody formation and clinical symptoms (AEs).
GP13-201
In study GP13-201, blood samples for ADA assessments were collected before the first infusion and at
Weeks 4, 16, 24, 38 and 52. If patients received a second treatment course, blood samples were again
collected before the first infusion and at the follow-up visit 26 weeks after the first infusion of the
second treatment course.
Across treatment arms, binding anti-rituximab antibodies were detected in 2 patients (1.2%) at
randomization (pre-treatment). These 2 patients were excluded from further ADA analysis.
The overall incidence of binding anti-rituximab antibodies in the Rixathon group was numerically lower
than that of the MabThera group. Overall, post-baseline ADA was detected in 9 patients (11%) in the
Rixathon and 18 patients (21.4%) in the MabThera group. There were no relevant differences observed
in terms of general safety between patients with- and without ADAs. Similarly, there was no
meaningful difference in efficacy outcome of the patients with- and without binding anti-rituximab
antibodies.
Assessment report
EMA/303207/2017 Page 96/115
Samples with confirmed positive binding anti-rituximab antibodies were further assessed for the
neutralizing capacity of the antibodies via a cell based assay (NAb assay). NAbs were detected in 3
patients in the Rixathon group (3.7%) compared to one patient in the MabThera group (1.2%). Among
them, two patients (1 in each arm) showed elevated ADA titer (> 100 ug/mL).
No clinically relevant differences could be observed for potentially hyper-acute or acute infusion
reactions or for potentially delayed reactions between both treatment groups.
There was no relevant difference in terms of efficacy between patients with- and without NAbs, as
shown by the DAS28 (CRP) profiles
GP13-301
In study GP13-301, blood samples for ADA assessments were collected at screening and at End of
Treatment of Combination and Maintenance Phase, respectively. For patients who participated in a
sparse pharmacokinetic sampling program, an additional sample was collected in Cycle 4 (pre-dose).
A total of 551 (87.6%) patients were included in the immunogenicity analysis, 268 (85.4%) patients in
the Rixathon group and 283 (89.8%) patients in the MabThera group. Five patients with pre-existing
anti-rituximab antibodies at screening visit were excluded from the immunogenicity assessment, 4
patients in the Rixathon group and 1 patient in the MabThera group. Of these 4 patients in the
Rixathon group, 3 patients were tested negative post-treatment, one patient was tested positive.
Overall, the number of patients with detectable post-dose anti-rituximab antibodies was low in the
study (8 out of 551 patients (1.5%)), with no clinically meaningful differences between the Rixathon
group (5 out of 268 patients, 1.9%) and the MabThera group (3 out of 283 patients, 1.1%). Based on
the limited PK data in ADA positive patients, there is no clear indication that ADA is having an impact
on PK exposure.
At the EOT in the Combination Phase, there were one Rixathon patient (0.4%) and two MabThera
patients (0.7%) who had confirmed positive results for binding-anti-rituximab antibodies.
At the EOT in the Maintenance Phase, there was only 1 patient (Rixathon) who was positive for binding
ADA.
No clinically relevant differences could be observed for potentially hyper-acute or acute infusion
reactions or for potentially delayed reactions between both treatment groups.
An assessment of safety parameters found that the four Rixathon patients who were tested ADA
positive during the Combination Phase had no AEs of infusion-related reactions.
Delayed hypersensitivity reactions have not been analyzed as AE of special interest in study GP13-301.
Therefore, only prominent potential AEs suspected to be related to study drug – like arthralgia,
myalgia, urticaria, skin rash, and pruritus which may form part of delayed hypersensitivity reactions –
can be compared without consideration of the time distance between the infusion and the occurrence
of the AEs.
Neutralizing antidrug-antibodies were detected in two out of 268 (0.7%) patients in the Rixathon group
and two out of 283 (0.7%) patients in the MabThera group.
NAb was detected in one of the patients in the Rixathon group at the end of Maintenance Phase.
All other NAb positive incidences were detected during the combination phase. These Nab incidences
showed no obvious link with any immunogenicity-related SAEs or diminished efficacy.
Assessment report
EMA/303207/2017 Page 97/115
Safety related to drug-drug interactions and other interactions
In the clinical development of Rixathon drug interactions have not been systematically investigated.
Rixathon was developed as a biosimilar medicinal drug product; therefore, the identified and potential
interactions of the reference product with other medicinal products MabThera as described in Section
4.5 of the MabThera SmPC also apply to Rixathon.
Discontinuation due to adverse events
GP13-201
The proportion of patients who prematurely discontinued study drug due to an AE was low for both
treatment groups (Rixathon: 4.7%, MabThera: 8.0%).
GP13-301
Combination Phase
During the Combination Phase of study GP13-301, the incidence of AEs leading to discontinuation of
study drug was similar for AEs of all grades between both treatment groups (Rixathon: 23 patients,
7.4%; MabThera: 22 patients, 7.0%). The incidence of Grade 3 or 4 AEs leading to discontinuation
study drug was comparable between both treatment groups (Rixathon: 14 patients [4.5%], MabThera:
12 patients [3.8%]).
Maintenance Phase
During the Combination Phase of study GP13-301, the incidence of AEs leading to discontinuation of
study drug was similar for AEs of all grades between both treatment groups (Rixathon: 23 patients,
7.4%; MabThera: 22 patients, 7.0%). The incidence of Grade 3 or 4 AEs leading to discontinuation
study drug was comparable between both treatment groups (Rixathon: 14 patients [4.5%], MabThera:
12 patients [3.8%]).
Post marketing experience
N/A
2.6.1. Discussion on clinical safety
The safety of Rixathon was evaluated in three clinical studies conducted in adult patients with RA
(study GP13-201), FL (study GP13-301) where MabThera was used as comparator, and other B-cell
NHLs (study GP13-101). The safety sample size is considered sufficient for a biosimilar application.
Study GP13-201
Although the Phase II study GP13-201 in RA patients was not powered to confirm safety, the outcomes
of this study are considered relevant especially as there is no other concurrent therapy.
The overall incidence of AEs was comparable between the treatment arms: (Rixathon: 56 patients,
65.1% vs. MabThera: 57 patients, 65.5%). The most commonly affected primary SOC were infections
and infestations, musculoskeletal and connective tissue disorders and gastrointestinal disorders. The
most common AEs included urinary tract infection, nasopharyngitis, (worsening of) rheumatoid
arthritis, hypertension, upper respiratory tract infection and bronchitis. The overall incidence of AEs
suspected to be related to study drug was similar for Rixathon (32.6%) and MabThera (33.3%) with
Assessment report
EMA/303207/2017 Page 98/115
urinary tract infection occurring more often in the Rixathon arm (5.8% [mostly mild, 1 moderate ] vs.
2.3%) and IRR, pruritus (each 1.2% vs. 4.6%) and rash (0 vs. 4.6%) occurring more often in the
MabThera arm.
The rate of discontinuation and dose adjustments/interruptions due to AEs was lower in the Rixathon
arm compared to the MabThera arm: 4.7% vs 8.0% and 7.0% vs. 12.6%, respectively.
No apparent differences in the frequency of SAEs were observed between the treatment groups:
Rixathon (14 patients, 6.1%), MabThera (10 patients, 4.3%). 1 death occurred in the Rixathon arm
which was attributed to an MTX overdose.
The incidence of AESI was similar between the treatment groups (Rixathon: 21 patients [24.4%];
MabThera: 22 patients [25.3%]). The most commonly reported AESI were rheumatoid arthritis (5.2%
overall), hypertension (4.6% overall) and nausea (4.0% overall).
Infusion-related reactions (IRR) were more commonly observed in the MabThera treatment arm
(Rixathon: 37.2%; MabThera: 42.5%). There were no notable differences in clinical laboratory
parameters.
Study GP13-301
In study GP13-301 as expected, the incidence of AEs including SAE (about 20%) during the pivotal
combination phase was higher than in the RA study, probably due to the more frequent rituximab
dosing regimen and the background therapy of CVP (cyclophosphamide-vincristine-prednisolone). E.g.
the higher incidence of nausea and neurological disorders is likely attributable to CVP. The overall
incidence of infections was similar in Rixathon (42.3%) and MabThera (41.9%) treatment groups.
The incidence of AEs was similar in both treatment groups (Rixathon -CVP: 92.6%; MabThera-CVP
91.4%). The most commonly affected primary SOCs were gastrointestinal disorders (primarily
constipation and nausea), nervous system disorders (primarily peripheral neuropathy, and
paraesthesia), infections and infestations (primarily urinary tract infection and upper respiratory tract
infection).
In the combination phase, neutropenia was far more commonly reported (Rixathon: 25.6%, MabThera:
29.5%), than in the RA study (1.2% for MabThera, no case for Rixathon). The rate of febrile
neutropenias was also higher in the Rixathon group (4.8% vs 2.9%). This is likely due to concurrent
cyclophosphamide treatment, but also the more intensive treatment of rituximab may further
contribute to neutropenia. E.g. in the original registration trial for the FL indication, neutropenia
occurred in 24% in the CVP+ rituximab arm, versus 13% in the CVP+ placebo arm. Notably, the
incidence of neutropenia in this study is similar to the historical data.
At least 5% absolute differences in AEs were observed for peripheral neuropathy and paresthesia.
However, differentiation of these terms may be difficult and when adding incidences for both terms
these differences between the treatment groups disappear.
The incidence of treatment discontinuations due to AEs was comparable between the treatment arms
(Rixathon: 23 patients, [7.4%]; MabThera: 22 patients, [7.0%]). The incidence of AEs requiring dose
interruption and/or reduction was lower in the Rixathon arm.
Adverse events of special interest included CRS and PML. None of these were observed during the
Combination Phase. The incidences of IRR considered to be related to study drug were comparable for
Rixathon and MabThera during Combination Phase (13.1% and 11.7%, respectively).
Assessment report
EMA/303207/2017 Page 99/115
Some adverse events which were observed more frequently in the Rixathon arm (urinary tract
infections, increased serum creatinine in GP13-201; cardiac disorders, ECG abnormalities in the
Combination Phase of study GP13-301) and especially during the Maintenance Phase of GP13-301
(higher incidences of AEs, Grade 4 AEs, SAEs, related AEs, discontinuation due to AEs, Neutropenia
and infections and infestations SOC) were further analysed and the findings from the updates balanced
out previous findings. The concern for neutropenia was resolved since it did not reflect in an increase
of infections and its evaluation of a carry-over effect of the earlier treatment combination phase with
chemotherapy (CVP regimen).
Study GP13-301 - Maintenance Phase
The incidence of AEs was higher in Rixathon group (63.2%) compared to the MabThera group (57.1%)
with injury, poisoning and procedural complications (Rixathon: 6.5%; MabThera: 3.0%) and vascular
disorders (Rixathon: 5.6%; MabThera: 3.5%) being the SOCs with the largest absolute difference of
3.5% and 2.1%, respectively, and neutropenia and cough being the more frequently observed AEs by
preferred term in the Rixathon group. More Grade 4 AEs were observed for the Rixathon treatment
group compared to MabThera (12 patients [5.2%] vs. 4 patients [1.7%]) again with neutropenia being
the most frequently observed AE. Treatment-related neutropenia occurred with a frequency > 5 %
(Rixathon, all grades 6.9%, grade 3/4 4.3%; MabThera, all grades 5.2%, grade 3/4 3.5%).
Overall, the percentage difference for AEs during the maintenance phase for the Rixathon arm
compared to the MabThera arm decreased from 6.1% to 1.6% with data cut-off from 10-Jul-2016.
The mechanism causing neutropenia following rituximab treatment is not well understood, as CD20 is
not expressed on neutrophil granulocytes. The mechanism of rituximab induced neutropenia has still
not been elucidated, while many hypotheses have been presented. Several authors suggest a
disturbance of haematopoiesis after rituximab treatment (Expert Rev Hematol. 2011;4(6):619-625).
As described in the SmPC of MabThera, neutropenia was not associated with a higher rate of infections
in earlier studies.
When CVP was withdrawn in the maintenance phase, the rate of neutropenia dropped considerably,
towards 10% (7.4 % grade 3/4) in the Rixathon arm, and 5.6% (3.9% Grade 3/4) in the MabThera
arm. The modest discrepancy between treatment arms is difficult to interpret. Reassuringly, this was
not associated with an enhanced rate of infections for Rixathon in the maintenance phase (cut –off 10-
Jul-2016: SOC infections and infestations - Rixathon: 32.7%; MabThera: 36.5%)).
The incidences of AEs leading to dose adjustment or interruption and AEs requiring additional therapy
were similar between the treatment groups. As for the slightly higher number of patients discontinuing
treatment due to an AEin the Rixathon arm, this percentage difference decreased with the data update
(10 [3.9%] in Rixathon and 7 [2.8%] in MabThera). There was a higher incidence of disease
progression in the Rixathon arm (N=37 [16.0%]) compared to the MabThera arm (N=25 [10.8%]).
However, the overall treatment discontinuation between the Rixathon arm and the MabThera arm is
reducing with the maturation of data.
Adverse events of special interest included CRS and PML. None of these were observed during the
Maintenance Phase. The incidences of IRR considered to be related to study drug were comparable for
Rixathon and MabThera during Maintenance Phase (10.8% and 8.2%, respectively).
Combination and Maintenance Phase – Deaths, Laboratory, Immunogenicity
The number of deaths occurring in the Combination, Maintenance and Post-treatment was comparable
between the treatment groups (Rixathon: 18 [5.8%]; MabThera: 17 [5.4%]).
Assessment report
EMA/303207/2017 Page 100/115
There were no notable differences in laboratory findings.
The number of ADA-positive patients was low for both treatment arms (Rixathon: 5 (1.9%) patients;
MabThera: 3 (1.1%) patients). NAbs were only detected in 4 patients (2 in each treatment arm).
Clinical data did not reveal a negative impact of ADA-positivity on the efficacy and safety of Rixathon.
Immunogenicity studies GP13-201 and GP13-301
In the RA study, the rate of ADA formation was higher for MabThera than for Rixathon (21.4% versus
11.0%). In many cases, the ADAs were transient. The number of neutralising ADA was small without
meaningful differences between groups (3.7% vs 1.2% for Rixathon and MabThera, respectively).
There was no clear relationship between ADAs and safety/efficacy. Moreover, based on quality assays,
no differences in immunogenicity are to be anticipated.
CVP including high dosages of prednisolone may suppress ADA formation more than methotrexate,
which was the background therapy in the RA study. Moreover, the dosages of rituximab were
considerable higher in the FL study in comparison to the RA study, which may further suppress
antibody formation. This may explain the low incidence of ADA of 1.5% in the combination phase. Also
in the maintenance phase the incidence of ADA remained low. However, as the data are yet incomplete
of this ongoing study, an update of ADA data is requested, also considering that the number of
subjects from the RA study was limited, and carry-over effect of CVP is expected to decrease in due
time (see RMP).
There were no relevant differences observed in terms of general safety between patients with and
without NAbs. No SAEs were reported for the NAb positive patients in the Rixathon arm, whereas two
suspected drug-related SAEs of moderate intensities (gastroenteritis and infusion reaction) were
reported for the patient in the MabThera arm. The study drug was permanently discontinued due to the
event. No relevant differences were observed in the DAS28 (CRP) profiles of ADA positive patients as
compared to the ADA negative patients. The DAS profiles of ADA positive and ADA negative patients
were equally distributed; no trending was observed which would suggest that ADA positivity leads to
either better or worse DAS28 response.
Immunogenicity results from study GP13-201 may indicate slight differences with respect to the ADA
incidence being 2-fold lower in RA patients treated with Rixathon compared to MabThera (11.0 % vs.
21.4 %), however, the data in the RA study were limited and at any rate, it is likely that the difference
of -10% could be a chance finding and ultimately it is not of concern since immunogenicity –if
anything—is lower than that of RMP which is in accordance with the overarching guidelines
(CHMP/437/04 Rev 1; EMEA/CHMP/BMWP/42832/2005 Rev1). In study GP13-301 there was no signal
of differences in immunogenicity and ADA formation between Rixathon (2.1%) and MabThera (0.9%).
The ADA formation may be suppressed by concurrent use of CVP in the main treatment phase of the
pivotal trial.
From the safety database of rituximab all the adverse reactions reported in clinical trials and post-
marketing have been included in the Summary of Product Characteristics which follows the one of
Mabthera.
2.6.2. Conclusions on the clinical safety
Overall, the safety profile of Rixathon appears to be comparable to the reference product MabThera
with most common reported events being infections, infusion related reactions, and in the NHL study,
neutropenia. In general, the frequencies and nature of the adverse events were similar between
Rixathon and MabThera, and in line with earlier reports for the Reference product MabThera in the RA
Assessment report
EMA/303207/2017 Page 101/115
and FL study populations. Final reports from the ongoing studies will provide further information on
clinical safety (see RMP).
2.7. Risk Management Plan
Safety concerns
Table 49 - Summary of the safety concerns
Important identified risks NHL/CLL Infusion-related reactions
Infections (including serious infections) Serious viral infections Impaired immunization response PML Neutropenia (including prolonged)
HBV reactivation Tumour lysis syndrome
GI perforation Stevens-Johnson Syndrome/ Toxic Epidermal Necrolysis RA Infusion-related reactions Infections (including serious infections) Impaired immunization response
PML Neutropenia (including prolonged) HBV reactivation Hypogammaglobulinemia Stevens-Johnson Syndrome/ Toxic Epidermal Necrolysis
GPA/MPA Infusion-related reactions Infections (including serious infections)
Impaired immunization response PML Neutropenia (including prolonged) HBV reactivation
Hypogammaglobulinemia Stevens-Johnson Syndrome/ Toxic Epidermal Necrolysis
Important potential risks NHL/CLL PRES Opportunistic infections Prolonged B-cell depletion
Increased risk of grade 3/4 serious blood and lymphatic system AEs in patients > 70 years (applicable for CLL only / relevant for marketing authorizations with indication CLL) AML/MDS Second malignancies Off-label use in pediatric patients
Administration route error RA PRES Opportunistic infections Malignant events Impact on cardiovascular disease
GI perforation
Assessment report
EMA/303207/2017 Page 102/115
Prolonged B-cell depletion Off-label use in autoimmune disease
Off-label use in pediatric patients GPA/MPA
PRES Opportunistic infections Malignant events Impact on cardiovascular disease GI perforation Prolonged B-cell depletion Off-label use in autoimmune disease
Off-label use in pediatric patients Relapses
Missing information NHL/CLL Use in Pregnancy and Lactation
RA Use in Pregnancy and Lactation Immunogenicity and autoimmune disease
GPA/MPA Use in Pregnancy and Lactation
Immunogenicity and autoimmune disease Long term use in GPA/MPA patients
Pharmacovigilance plan
Table 50 - On-going and planned additional PhV studies/activities in the Pharmacovigilance
Plan
Study/activity
(including
study
number)
Objectives Safety concerns
/efficacy issue
addressed
Status Planned date for
submission of (interim
and) final results
GP13-201
(Study Part 2): A randomized, double-blind, controlled study to evaluate pharmacokinetic
s, pharmacodynamics, safety and efficacy of Rixathon and rituximab in
patients with rheumatoid arthritis
refractory or intolerant to standard DMARDs and
one or up to three anti-TNF therapies
Primary
objective: To assess bioequivalence between Rixathon and rituximab
Safety objectives:
- Overall safety
and tolerability
- Incidence of
anti-drug antibodies
Immunogenicity, infusion related reactions, infections
(including serious
infections)
Ongoing Part I:
Will be part of the
submission for marketing authorization in Apr 2016
Part II
24 week interim report: Dec 2016 (finalized)
52 week final report: Nov
2017
Assessment report
EMA/303207/2017 Page 103/115
Study/activity
(including
study
number)
Objectives Safety concerns
/efficacy issue
addressed
Status Planned date for
submission of (interim
and) final results
Category 3
GP13-301: A
randomized, controlled, double-blind Phase III trial to compare the
efficacy, safety and pharmacokinetics of Rixathon plus cyclophosphami
de, vincristine, prednisone vs. MabThera® plus cyclophosphamide, vincristine, prednisone, followed by
Rixathon or MabThera® maintenance therapy in patients with previously untreated,
advanced stage follicular lymphoma
Category 3
Primary objective
To demonstrate
comparability of the overall response rate (ORR)
Safety objective
- Safety of
Rixathon in
comparison to MabThera® either
as single agent or in combination with CVP
- Incidence of
immunogenicity (anti-drug antibody
formation)
Infusion related reactions, infections
(including serious infections), serious viral infections, neutropenia (including prolonged
neutropenia), opportunistic
infections
Ongoing Combination phase (cut-
off date 10 Jul 2015):
Will be part of the
submission for marketing authorization in Apr 2016
Interims Analysis (cut-off
date 10 Jul 2016): Dec 2016 (finalized)
Final report: Aug 2018
GP13-302: A
randomized, double- blind, controlled, parallel-group, multicenter study to assess the safety and
immunogenicity of transitioning to Rixathon or re-treatment with Rituxan® or MabThera® in
patients with
active rheumatoid arthritis, previously treated with Rituxan® or
MabThera®
Identify potential
safety risk of the transition from reference product to Rixathon as compared to continuing with respective
treatment weight
Immunogenicity, acute infusion related reactions, infections (including serious
viral infections)
Ongoing 12 week interim report:
Feb 2017
24 week report: Jul 2017
Assessment report
EMA/303207/2017 Page 104/115
Study/activity
(including
study
number)
Objectives Safety concerns
/efficacy issue
addressed
Status Planned date for
submission of (interim
and) final results
Category 3
British Society
of Rheumatology Biologics Register
(BSRBR),
Swedish registry
(ARTIS)
German registry
(RABBIT)
Category 3
Provide additional
supporting safety data in RA to further characterize the
nature of events, demographics of
patients at risk, and the presence of risk factors and confounding factors.
For all 3
registries: Infusion-related reactions, Infections
(including serious infections),
Impaired immunization response, PML, Neutropenia (including
prolonged), HBV reactivation, Posterior reversible encephalopathy syndrome
(PRES), Opportunistic infections, Malignant events, Impact on cardiovascular
disease, GI
perforation, Prolonged B-cell depletion, Immunogenicity and autoimmune disease
Planned, (Start at time of
drug availability in country
following
EMA
approval )
Interim report planned yearly in Q4 for 4 years
starting in 2018, final report within 6-12 months after study
completion
Risk minimisation measures
Table 51 - Summary table of Risk Minimization Measures
Safety concern
Routine risk
minimization measures
Additional risk
minimization measures
Infusion-related reactions
For NHL/CLL: 4.2, 4.4, 4.8 of
the SmPC
For RA: 4.2, 4.4, 4.8, 5.1 of
the SmPC
HCP educational leaflet
For GPA/MPA: 4.2, 4.4, 4.8 of
the SmPC
HCP educational leaflet
Assessment report
EMA/303207/2017 Page 105/115
Safety concern
Routine risk
minimization measures
Additional risk
minimization measures
Infections (including serious
infections)
For NHL/CLL, GPA/MPA: 4.3,
4.4, 4.8 of the SmPC
GPA/MPA: HCP educational
leaflet, Patient educational
leaflet, Patient alert card
For RA: 4.3, 4.4, 4.5, 4.8 of
the SmPC
HCP educational leaflet,
Patient educational leaflet,
Patient alert card
Serious viral infections For NHL/CLL: 4.3, 4.4, 4.8 of
the SmPC
None
Impaired immunization
response
For NHL/CLL, RA, GPA/MPA:
4.4 of the SmPC
None
PML
For NHL/CLL: 4.3, 4.4, 4.8 of
the SmPC
None
For RA: 4.3, 4.4, 4.8, 5.1 of
the SmPC
HCP educational leaflet,
Patient educational leaflet,
Patient alert card
For GPA/MPA: 4.3, 4.4 of the
SmPC
HCP educational leaflet,
Patient educational leaflet,
Patient alert card
Neutropenia (including
prolonged)
For NHL/CLL, RA, MPA/GPA:
4.4, 4.8 of the SmPC
None
HBV reactivation
For NHL/CLL, RA, MPA/GPA:
4.4, 4.8 of the SmPC
None
Tumour lysis syndrome For NHL/CLL: 4.2, 4.4, 4.8 of
the SmPC
None
GI perforation
For NHL/CLL, RA, MPA/GPA:
4.8 of the SmPC
None
Stevens-Johnson Syndrome/
Toxic Epidermal Necrolysis
For NHL/CLL, RA, MPA/GPA:
4.4, 4.8 of the SmPC
None
Hypogammaglobulinemia
For RA, MPA/GPA: 4.4, 4.8 of
the SmPC
None
PRES For NHL/CLL, RA, GPA/MPA:
4.8 of the SmPC
None
Assessment report
EMA/303207/2017 Page 106/115
Safety concern
Routine risk
minimization measures
Additional risk
minimization measures
Opportunistic infections
For NHL/CLL, GPA/MPA: 4.3,
4.4, 4.8 of the SmPC
None
For RA: 4.3, 4.4, 4.5, 4.8 of
the SmPC
None
Prolonged B-cell depletion
For NHL/CLL, RA: 4.8, 5.1 of
the SmPC
None
For GPA/MPA: 5.1 of the SmPC None
Increased risk of grade 3/4
serious blood and lymphatic
system AEs in patients > 70
years (applicable for CLL only
/ relevant for marketing
authorizations with indication
CLL)
For CLL: 4.8 of the SmPC None
AML/MDS For NHL/CLL: 4.4 of the SmPC None
Second malignancies For NHL/CLL: 4.4 of the SmPC None
Off-label use in pediatric
patients
For NHL/CLL, RA: 4.2 of the
SmPC
None
For GPA/MPA: 4.1, 4.2 of the
SmPC
None
Administration route error For NHL/CLL: 4.2 of the SmPC
The outer carton as well as the
vial label of the product states:
For intravenous use after
dilution.
HCP educational leaflet
Malignant events
For RA: 4.4 5.1 of the SmPC None
For GPA/MPA: 4.8 of the SmPC None
Impact on cardiovascular
disease
For RA, GPA/MPA: 4.2, 4.3,
4.4, 4.8 of the SmPC
None
Off-label use in autoimmune
disease
For RA, GPA/MPA: 4.8 of the
SmPC
None
Relapses For GPA/MPA: Currently
available data do not support
the need for risk minimization.
None
Use in Pregnancy and For NHL/CLL, RA, GPA/MPA: None
Assessment report
EMA/303207/2017 Page 107/115
Safety concern
Routine risk
minimization measures
Additional risk
minimization measures
Lactation
4.6, 5.3 of the SmPC
Immunogenicity and
autoimmune disease
For RA, GPA/MPA: 4.5, 5.1 of
the SmPC
None
Long term use in GPA/MPA
patients
For GPA/MPA: Currently
available data do not support
the need for risk minimization.
None
Conclusion
The CHMP and PRAC considered that the risk management plan version 1.4 is acceptable.
2.8. Pharmacovigilance
Pharmacovigilance system
The CHMP considered that the pharmacovigilance system summary submitted by the applicant fulfils
the requirements of Article 8(3) of Directive 2001/83/EC.
2.9. Product information
2.9.1. User consultation
The results of the user consultation with target patient groups on the package leaflet submitted by the
applicant show that the package leaflet meets the criteria for readability as set out in the Guideline on
the readability of the label and package leaflet of medicinal products for human use. No full user
consultation with target patient groups on the package leaflet has been performed on the basis of a
bridging report making reference to MabThera. The bridging report submitted by the applicant has
been found acceptable.
2.9.2. Additional monitoring
Pursuant to Article 23(1) of Regulation No (EU) 726/2004, Rixathon (rituximab) is included in the
additional monitoring list as it is a biological product authorised after 1 January 2011.
Therefore the summary of product characteristics and the package leaflet includes a statement that
this medicinal product is subject to additional monitoring and that this will allow quick identification of
new safety information. The statement is preceded by an inverted equilateral black triangle
Assessment report
EMA/303207/2017 Page 108/115
3. Benefit-Risk Balance
3.1. Therapeutic Context
The therapeutic context of rituximab is very well described over the years since it first received the MA
in the EU (2nd June 1998), as Mabthera.
Rixathon (rituximab) has been developed as a biosimilar of MabThera, the reference product.
3.1.1. Disease or condition
The approval is sought for all approved indications of the reference product MabThera in the EU,
according to the MabThera Summary of Product Characteristics (SmPC); Non-Hodgkin’s lymphoma
(NHL: diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL)); Chronic lymphocytic
leukemia (CLL); Rheumatoid arthritis (RA) and Granulomatosis with polyangiitis (GPA) and microscopic
polyangiitis (MPA).
3.1.2. Available therapies and unmet medical need
The originator MabThera is available in the EU since 1998; therefore there is no unmet medical need.
3.1.3. Main clinical studies
A pivotal PK/PD study was performed in 173 patients with RA who were irresponsive or intolerant to
one or more TNF-alpha inhibitors, comparing a standard treatment course of either Rixathon to
MabThera of 1,000 mg per iv infusion, two weeks apart. After 24 weeks, patients could be re-treated
based on the presence of residual disease activity (DAS28>2.6), in line with the approved posology of
MabThera. The study was powered to assess equivalence of PK in terms of the area under the serum
concentration-time curve from time zero to infinity (AUC(0-inf)) and PD in terms of B-cell depletion
between Rixathon and MabThera (GP13- 201 Study Part I). The study also provides data on key
efficacy and safety including immunogenicity variables.
A pivotal efficacy and safety study in 627 patients with previously untreated advanced FL (study GP13-
301). In total 627 subjects received at least one study treatment. Patients were randomised to either
Rixathon or MabThera 375 mg/m2 IV for 8 cycles (every 3 weeks), in combination with CVP
(cyclophosphamide, vincristine, prednisone). The study was powered to demonstrate equivalence of
clinical response (ORR at week 24) as the primary endpoint between Rixathon and MabThera
(equivalence margin -/+ 12%) and to provide data on PK/PD and safety including immunogenicity
assessments. Primary analyses were performed in the Per-Protocol Set (n=624). PK-PD data were
obtained from a small subset of 48 subjects. The study consisted of a Combination Treatment Phase
followed by a Maintenance Treatment Phase in which responding patients continued earlier assigned
treatment with Rixathon or MabThera as monotherapy (375 mg/m2 every 2-3 months) for another 24
months (n=462). This part of the study is still ongoing at the time of data cut-off of for this MAA.
3.2. Favourable effects
Since the favourable effects of rituximab (as Mabthera) are well established, this application aims to
prove similarity between Rixathon (Rixathon) and the originator MabThera. This is confirmed on non-
Assessment report
EMA/303207/2017 Page 109/115
clinical grounds with regard to: antibody Dependent Cellular Cytotoxicity (ADCC) in different assay set-
ups; B-cell depletion in cynomolgus monkeys after repeated dosing of 20 or 100 mg/kg 4qw; survival
in xenografted mouse models; tumour growth in xenografted mouse models.
The required bioequivalence criterion of 90% CI of 80 - 125% between Rixathon and MabThera for the
primary parameter AUC0-inf was fulfilled and this was confirmed by a sensitivity analysis including
body surface area as an additional covariate in the pivotal study GP13-201 performed in RA patients.
In Study GP13-201, the ratio of the geometric means of AUEC(0-14d) between Rixathon and MabThera
was 1.019 (95% CI 0.997, 1.042), which is well within the pre-set bioequivalence margin of 0.80-1.25.
This is further supported by exploratory analyses in the two NHL studies, showing a rapid induction
and persistent low level of B-cells during treatment of GP-2012, similar to MabThera.
In study GP13-201 the key secondary efficacy endpoint was change from baseline in DAS28 (CRP) at
Week 24. The criterion for non-inferiority was met. Results from the ACR20 response analysis at week
24 and averaged ACR20 responders between week 4 and week 24 showed that Rixathon and MabThera
are similar regarding ACR20 rates with when applying equivalence margins of ±15%.
In the pivotal efficacy and safety study GP13-301 equivalence with regards to ORR (primary efficacy
endpoint) was demonstrated as the entire 95% CI for the difference in ORR between the two
treatments was within the pre-specified equivalence margin of ±12%. BOR was assessed as secondary
efficacy endpoint. The proportions of patients in Combination Treatment Phase with the best overall
response (BOR) based on central blinded review of tumour assessments (CR, PR, stable disease and
progressive disease) and their associated 90% CIs were similar for the four tumour assessment
categories.
3.3. Uncertainties and limitations about favourable effects
There are no uncertainties regarding the biosimilarity of Rixathon to MabTthera in terms of the
pharmacokinetic and pharmacodynamic profile.
In study GP13-301 at (data cut-off: 31-Dec-2016) more patients in the MabThera arm are on ongoing
treatment whereas a higher number of patients treated with Rixathon than MabThera ended treatment
in the maintenance phase with the primary reason for discontinuation being disease progression
(20.9% versus 14.3%). The HR for PFS (Rixathon/MabThera) was 1.31 (90% CI [1.02, 1.69]), at the
December cut-off, in the same range as observed with the first PFS analysis (data cut off: 10-Jul-
2016) where the PFS HR was calculated to be 1.25 (90% CI: [0.96, 1.61]). However, as study GP13-
301 was not powered for time-to-event outcomes, hence, for PFS and OS the currently observed data
are still immature. Moreover, the follow-up time up to now is too short to allow for an estimation of
median PFS and the number of PFS events low and the rate of censoring high. The availability of the
study report will provide further information on PFS (see RMP).
3.4. Unfavourable effects
The safety profile of Rixathon in study GP13-201 was comparable to MabThera with regards to the
incidence of overall adverse events, SAEs and AESI. The following incidences were even lower in the
Rixathon arm: premature discontinuation and dose adjustments/interruptions due to AEs, Infusion-
related reactions and overall ADA.
Infections and infusion-related reactions were the most commonly reported events (RA Study GP13-
201: Rixathon: 31.4 and 37.2%, respectively; MabThera: 35.6% and 42.5%, respectively).
Assessment report
EMA/303207/2017 Page 110/115
In study GP13-301 during the Combination Phase the safety profile of Rixathon was comparable to
MabThera with regards to the incidence of overall adverse events, Grade 3-4 adverse events,
suspected treatment related adverse events, treatment discontinuation due to AEs, dose adjustment or
interruption and AEs requiring additional therapy as well as deaths.
The incidences of IRR considered to be related to study drug were comparable for Rixathon and
MabThera during Combination Phase.
Overall, in the Maintenance Phase, the following incidences were higher in the Rixathon group: Overall
Adverse events, Grade 4 AEs, suspected treatment-related AEs, treatment discontinuation due to AEs,
serious adverse events, Neutropenia all Grades, Infections and infestation most frequent SOC in
serious adverse events and reason for treatment discontinuation. With the data update from the cut-
off from 10 July 2016 the percentage difference for AEs during the maintenance phase for the Rixathon
arm compared to the MabThera arm decreased from 6.1% to 1.6%. The percentage number of SAEs
and discontinuation due to AEs was comparable between the treatment arms. Although, the number of
AEs suspected to be related to study drug is slightly higher for the Rixathon arm, the number of Grade
3-4 AEs as well as the number of SAEs are comparable between the treatment arms.
There were no meaningful differences in ADA formation between Rixathon and MabThera. In the RA
study, ADAs were detected in 11.0% versus 21.4% in the Rixathon and MabThera arm, respectively.
Neutralising ADAs were detected in 3.7% of the Rixathon arm and 1.2% of the MabThera arm. There
was no clear relationship between the presence of ADA’s and efficacy/safety. Overall, the number of
patients with ADA was low in Study GP13-301 in Follicular Lymphoma (8 /551 subjects, 1.5%), without
meaningful differences between treatment assignments. The rate of ADA formation may be reduced by
concurrent chemotherapy or prednisolone.
In order to further investigate the risks of Immunogenicity, infusion related reactions, infections
(including serious infections), neutropenia, the applicant will conduct and submit the results of:
- GP13-201 (Study Part 2), a randomised, double-blind, controlled study to evaluate pharmacokinetics,
pharmacodynamics, safety and efficacy of Rixathon and rituximab in patients with rheumatoid arthritis
refractory or intolerant to standard DMARDs and one or up to three anti-TNF therapies.
- GP13-301, a randomised, controlled, double-blind Phase III trial to compare the efficacy, safety and
pharmacokinetics of Rixathon plus cyclophosphamide, vincristine, prednisone vs. MabThera plus
cyclophosphamide, vincristine, prednisone, followed by Rixathon or MabThera maintenance therapy in
patients with previously untreated, advanced stage follicular lymphomain patients with previously
untreated, advanced stage follicular lymphoma.
- GP13-302: A randomized, double- blind, controlled, parallel-group, multicenter study to assess the
safety and immunogenicity of transitioning to Rixathon or re-treatment with Rituxan / Riximyo or
MabThera in patients with active rheumatoid arthritis, previously treated with Rituxan / Riximyo or
MabThera.
- Data deriving from the following registries: British Society of Rheumatology Biologics Register
(BSRBR), Swedish registry (ARTIS) and German registry (RABBIT).
Please see RMP section 2.7.
3.5. Uncertainties and limitations about unfavourable effects
There are no uncertainties concerning the comparability of the clinical safety of Rixathon with
Mabthera. Further long term safety information will be provided with the final study report (see RMP).
Assessment report
EMA/303207/2017 Page 111/115
3.6. Effects Table
Not applicable for biosimilars.
3.7. Benefit-risk assessment and discussion
3.7.1. Importance of favourable and unfavourable effects
A comprehensive biosimilarity exercise, which covered all relevant structural and functional
characteristics of the rituximab molecule, was submitted. The presented results support the
biosimilarity claim; similarity between Rixathon and the EU reference product MabThera is considered
demonstrated at the quality level. Any minor differences observed have been adequately justified with
respect to the efficacy/safety profile of Rixathon.
Comparability data on three relevant mechanisms of action (CDC, ADCC, ADCP and apoptosis), which
are considered important for lysis of the B-cell subsequent to binding of rituximab to the CD20 antigen
were provided. Bioequivalence was demonstrated for the primary PK endpoint AUC(0-inf) as well as
mostly all other secondary PK endpoints. The descriptive data for the PK parameters determined in
study GP13-201 indicate comparability between Rixathon and MabThera.
The key secondary efficacy endpoint in study GP13-201 was met and results of the analyses for the
ACR20 rates of response at week 24 and averaged ACR20 responders between week 4 and week 24
were within the applied equivalence margins of ±15% thereby suggesting similarity.
The primary efficacy endpoint (ORR) in GP13-301 was met. Updated OS and PFS data will be
submitted with the final study report (see RMP).
The adverse events with Rixathon were in overall in line with the well-established safety profile of
Mabthera. Overall, the rates of ADA formation were low in both the RA trial and the study in FL.
Updates on safety will be submitted with the final CSRs of the ongoing studies (see RMP).
3.7.2. Balance of benefits and risks
Biosimilarity of Rixathon to the originator Mabthera has been demonstrated with regards to PK/PD,
efficacy and safety parameters in two clinical trials and two different indications Rheumatoid Arthritis
and Follicular Lymphoma.
3.7.3. Additional considerations on the benefit-risk balance
With regards to the efficacy, it is well established that the mechanism of action and PD aspects are
common across autoimmune and across oncology indications of Mabthera. Therefore, and in line with
the EMA guidelines on the similar biological medicinal products, the efficacy results obtained with
Rixathon, demonstrating equivalence with Mabthera in RA and FL patients can be reasonably
extrapolated to the other approved therapeutic indications of Mabthera.
The applicant claims the same therapeutic indications for adult patients for the biosimilar Rixathon as
granted for Mabthera for iv administration in the EU. However, as Mabthera is also marketed in the
subcutaneous indication, a risk of medication error has been identified. Adequate risk minimisation
measures to avoid the potential route of administration error have been included in the RMP.
Assessment report
EMA/303207/2017 Page 112/115
3.8. Conclusions
Rixathon is considered biosimilar to Mabthera and therefore the overall Benefit Risk balance of
Rixathon is considered positive in the following indications:
Non-Hodgkin’s lymphoma (NHL)
Rixathon is indicated for the treatment of previously untreated patients with stage III-IV follicular
lymphoma in combination with chemotherapy.
Rixathon maintenance therapy is indicated for the treatment of follicular lymphoma patients
responding to induction therapy.
Rixathon monotherapy is indicated for treatment of patients with stage III-IV follicular lymphoma who
are chemo-resistant or are in their second or subsequent relapse after chemotherapy.
Rixathon is indicated for the treatment of patients with CD20 positive diffuse large B cell non-Hodgkin’s
lymphoma in combination with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisolone)
chemotherapy.
Chronic lymphocytic leukaemia (CLL)
Rixathon in combination with chemotherapy is indicated for the treatment of patients with previously
untreated and relapsed/refractory chronic lymphocytic leukaemia. Only limited data are available on
efficacy and safety for patients previously treated with monoclonal antibodies including Rixathon or
patients refractory to previous Rixathon plus chemotherapy.
Rheumatoid arthritis
Rixathon in combination with methotrexate is indicated for the treatment of adult patients with severe
active rheumatoid arthritis who have had an inadequate response or intolerance to other disease-
modifying anti-rheumatic drugs (DMARD) including one or more tumour necrosis factor (TNF) inhibitor
therapies.
Rixathon has been shown to reduce the rate of progression of joint damage as measured by X-ray and
to improve physical function, when given in combination with methotrexate.
Granulomatosis with polyangiitis and microscopic polyangiitis
Rixathon, in combination with glucocorticoids, is indicated for the induction of remission in adult
patients with severe, active granulomatosis with polyangiitis (Wegener’s) (GPA) and microscopic
polyangiitis (MPA).
4. Recommendations
Similarity with authorised orphan medicinal products
The CHMP by consensus is of the opinion that Rixathon is not similar to Imbruvica (Ibrutinib), Arzerra
(Ofatuzumab), Gazyvaro (Obinutuzumab) and Venclyxto (Venetoclax) within the meaning of Article 3
of Commission Regulation (EC) No. 847/200. See appendix 1.
Outcome
Based on the CHMP review of data on quality, safety and efficacy, the CHMP considers by consensus
that the risk-benefit balance of Rixathon is favourable in the following indications:
Rixathon is indicated in adults for the following indications:
Assessment report
EMA/303207/2017 Page 113/115
Non-Hodgkin’s lymphoma (NHL)
Rixathon is indicated for the treatment of previously untreated patients with stage III IV follicular
lymphoma in combination with chemotherapy.
Rixathon maintenance therapy is indicated for the treatment of follicular lymphoma patients
responding to induction therapy.
Rixathon monotherapy is indicated for treatment of patients with stage III IV follicular lymphoma who
are chemo-resistant or are in their second or subsequent relapse after chemotherapy.
Rixathon is indicated for the treatment of patients with CD20 positive diffuse large B cell non Hodgkin’s
lymphoma in combination with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisolone)
chemotherapy.
Chronic lymphocytic leukaemia (CLL)
Rixathon in combination with chemotherapy is indicated for the treatment of patients with previously
untreated and relapsed/refractory CLL. Only limited data are available on efficacy and safety for
patients previously treated with monoclonal antibodies including Rixathon or patients refractory to
previous Rixathon plus chemotherapy.
See section 5.1 for further information.
Rheumatoid arthritis
Rixathon in combination with methotrexate is indicated for the treatment of adult patients with severe
active rheumatoid arthritis who have had an inadequate response or intolerance to other disease
modifying anti rheumatic drugs (DMARD) including one or more tumour necrosis factor (TNF) inhibitor
therapies.
Rixathon has been shown to reduce the rate of progression of joint damage as measured by X ray and
to improve physical function, when given in combination with methotrexate.
Granulomatosis with polyangiitis and microscopic polyangiitis
Rixathon, in combination with glucocorticoids, is indicated for the induction of remission in adult
patients with severe, active granulomatosis with polyangiitis (Wegener’s) (GPA) and microscopic
polyangiitis (MPA).
The CHMP therefore recommends the granting of the marketing authorisation subject to the following
conditions:
Conditions or restrictions regarding supply and use
Medicinal product subject to restricted medical prescription (see Annex I: Summary of Product
Characteristics, section 4.2).
Other conditions and requirements of the marketing authorisation
Periodic Safety Update Reports
The requirements for submission of periodic safety update reports for this medicinal product are set
out in the list of Union reference dates (EURD list) provided for under Article 107c(7) of Directive
2001/83/EC and any subsequent updates published on the European medicines web-portal.
Assessment report
EMA/303207/2017 Page 114/115
Conditions or restrictions with regard to the safe and effective use of the medicinal product
Risk Management Plan (RMP)
The MAH shall perform the required pharmacovigilance activities and interventions detailed in the
agreed RMP presented in Module 1.8.2 of the marketing authorisation and any agreed subsequent
updates of the RMP.
An updated RMP should be submitted:
At the request of the European Medicines Agency;
Whenever the risk management system is modified, especially as the result of new
information being received that may lead to a significant change to the benefit/risk profile or
as the result of an important (pharmacovigilance or risk minimisation) milestone being
reached.
Additional risk minimisation measures
Non-oncology indications:
The MAH must ensure that all physicians who are expected to prescribe Rixathon are provided with the
following:
Product information
Physician information
Patient information
Patient Alert card
The Physician information about Rixathon should contain the following key elements:
• The need for close supervision during administration in an environment where full resuscitation
facilities are immediately available
• The need to check, prior to Rixathon treatment, for infections, for immunosuppression, for
prior/current medication affecting the immune system and recent history of, or planned,
vaccination
• The need to monitor patients for infections, especially PML, during and after Rixathon
treatment
• Detailed information on the risk of PML, the need for timely diagnosis of PML and appropriate
measures to diagnose PML
• The need to advise patients on the risk of infections and PML, including the symptoms to be
aware of and the need to contact their doctor immediately if they experience any.
• The need to provide patients with the Patient Alert Card with each infusion
The Patient information about Rixathon should contain the following key elements:
• Detailed information on the risk of infections and PML
• Information on the signs and symptoms of infections, especially PML, and the need to contact
their doctor immediately if they experience any
• The importance of sharing this information with their partner or caregiver
Assessment report
EMA/303207/2017 Page 115/115
• Information on the Patient Alert Card
The Patient Alert Card for Rixathon in non-oncology indications should contain the following key
elements:
• The need to carry the card at all times and to show the card to all treating health care
professionals
• Warning on the risk of infections and PML, including the symptoms
• The need for patients to contact their health care professional if symptoms occur
Oncology indications:
The MAH must ensure that all physicians who are expected to prescribe Rixathon are provided with the
following:
Product information
Physician information
The Physician information about Rixathon should contain the following key elements:
Information that the product should be administered as IV only to avoid administration route
errors.
The Physician information and Patient information must be agreed with the National Competent
Authorities prior to distribution and Patient Alert Card should be included as part of inner packaging.